[00:22] Argh! Why is compiz claiming +? for alt-tab? [00:23] *Sometimes* ? [00:30] RAOF: I am having other funky keybinding issues in conjunction with Orca. [00:30] So I think there may be keybinding issues with recent compizz. [00:31] I think this might have to do with the recent keybindings rework. :) [00:31] RAOF: Indeed. [00:31] I've just filed a bug for my issue. [00:51] Huh. It seems that the *second* keypress works - is swallowed, but , goes back a page in Chromium. [00:52] Heh. And if you , Tomboy actually gets the ! Workaround! [01:28] * RAOF kills compiz again to get his keyboard back. [01:29] …and again. [02:25] @pilot out [05:47] Good morning [05:48] dobey: we don't have many languages any more; English, Simplified Chinese (for China), Spanish, Portugese [05:49] all others already became a victim of adding the 25th runtime/language on the CD [06:00] RAOF, sorry, had to reject your branch [06:01] DBO: The pointer-barrier thing? Yeah, after using it for a while it turns out that having the barrier both ways *is* better ;) [06:02] RAOF: are you running unity-team ppa for compiz? [06:02] jasoncwarner_: Yes, I think so. [06:02] Oh, actually no. [06:02] jasoncwarner_, I am [06:03] if you are referencing evil things from the super key-presses, we know [06:03] RAOF: thumper just told me that the compiz there has some serious key grabbing issues...so if you are, that could be a problem [06:03] DBO: yeah, thumper just walked me through it... [06:03] I *am* running unity-team PPA for unity, though. [06:03] RAOF: I think you would have the compiz as well then, no? [06:04] DBO: thanks, yo...sounds like that one is going to be a fun one to look into ;) [06:04] 1:0.9.7.0~bzr2995-0ubuntu3 is the version of compiz I have; it's from the main archive. [06:04] I also have fun keygrabbing issues; is it possible that the compiz with that issue is in the main archive, rather than the PPA? [06:04] RAOF: yeah, me too Installed: 1:0.9.7.0~bzr2995-0ubuntu3 [06:04] RAOF: I'm pretty sure that is unity-team ppa (didrocks can confirm, though) [06:04] RAOF, what keygrabbing issues are you seeing? [06:05] DBO: brings up alt-tab switcher, doesn't work, but does. [06:05] Similarly, doesn't summon Do; does. [06:06] works for [06:06] alt right? [06:06] Basically, it seems that unity is eating the first keypress of any modifier-using chord. [06:07] RAOF, how could we be doing that? [06:07] DBO: For getting dot points in Tomboy, for example, or for forward in Chromium. [06:07] super space isn't even us [06:07] dobey: oh, there's one other thing: if we can drop python3 from the default install again, we'd reclaim some 5 MB :) [06:07] DBO: No, but you *could* be eating everything with a attached, again. You've got a passive grab on both and now, right? [06:07] RAOF, also we got unity running in LLVM pipe [06:08] Except now you're only eating the first keypress rather than all of them :) [06:08] RAOF, nothing in archive should have changed that [06:09] RAOF, LLVMpipe unity is quite fast [06:09] There is no newer compiz in the unity-team PPA, at least as far as I can see. [06:09] DBO: Yeah, LLVMpipe is pretty good. I'd be interested to see how it fares on less-than-awesome CPUs; ARM would be a nice test :) [06:11] RAOF, works fine on a mini-9 [06:11] That's pretty impressive. [06:11] yeah [06:11] its not *as* fast [06:11] but its livable [06:12] Slowish when trying to render a blurred alt-tab over video, I'd wager, but otherwise fine? [06:12] yes [06:12] llvm-pipe should disable blurs [06:12] otherwise it is plenty fast [06:12] It just required you to swapbuffers on each damage event? [06:12] yes [06:12] slash frame? [06:12] yes [06:13] which we cna do from the unity plugin itself [06:13] so, as far as Unity is concerned [06:13] we run on everything now [06:13] Props. [06:14] Now to get didrocks to try that barrier stuff :) [06:15] Hey! Why don't I have software-center installed? [06:18] DBO: The patch for https://bugs.launchpad.net/compiz-core/+bug/925293 was in 1:0.9.7.0~bzr2995-0ubuntu2, which is in the archive. Is that the thing causing modifiers to go mad? [06:18] Launchpad bug 925293 in compiz-core "Plugins can't tell the difference between a modifier key-tap, and a modifier key-release (after being used to modify other keys)" [High,Fix committed] [06:19] good morning [06:19] Yo didrocks! [06:19] hey RAOF [06:19] RAOF, yeah thats the one [06:19] okay [06:19] modifier going mad? [06:20] hey didrocks [06:20] guten morgen pitti [06:20] didrocks: Press :) [06:20] RAOF, DBO: oh, are you talking about the problem that I can't type "B" any more? [06:20] RAOF: works here, weechat changes my channel [06:20] Alternatively, try to go back a page in firefox by pressing [06:20] (I have a Ctrl+B keybinding assigned which is broken now, and then couldn't type B without modifier) [06:20] works too [06:21] didrocks: What's your compiz? [06:21] RAOF: well, the latest I pushed to the archive [06:21] ubuntu3 [06:21] Hm, odd. [06:22] RAOF: pitti: do you have other examples of broken keys? [06:22] * didrocks already reverted one patch yesterday from -0ubuntu2, so I can revert another one :( [06:22] didrocks: Anything using or . The first keypress is ignored (or does something weird), the second and subsequent has the normal effect. [06:23] So I *can* go back a page in Chromium by hitting , but does nothing. [06:23] pitti, yeah [06:23] pitti, seems to be the patch to compiz that got backported into distro [06:23] Also, no longer summons Do; does. [06:23] grrr, I don't understand, it works here… [06:23] The bane of the developer! [06:24] ok, anyway, if there is a patch causing that, it can only be this one from the current set [06:24] hum [06:24] no, there is maybe the keybinding one as well [06:25] anyway, that patch doesn't seem to have helped here to repair broken key bindings, only made it worse :/ [06:25] pitti: it's not that one [06:25] or putting windows into the background is generally broken now [06:25] pitti: it's another one which repairs the keybinding [06:25] pitti: so, open ccsm [06:25] pitti: viewport [06:25] switch to viewport [06:25] then assign a keybinding [06:25] -> this should work [06:26] hm, my broken key is with window management [06:26] (which are in control-center) [06:27] I don't see a key for "put into background" in ccsm, but I'll keep lokoing [06:27] pitti: hum, I'm really not sure about this one though [06:27] is the viewport keybinding working? [06:27] that is, switch between virtual desktops? [06:27] yes, that never broke [06:27] (people were complaining about this one for keybindings [06:28] it was broken [06:28] in fact, all of the Alt+F5, Alt+F10, Strg+Alt+Cursor etc. kept working all the time for me [06:28] except my Ctrl+B to put window into the background [06:28] pitti: directly switching to a viewport [06:28] but I guess I ought to file a bug about it [06:28] didrocks: ah, let me try [06:28] viewport switcher -> go to specific viewport [06:28] people were complaining about those [06:28] and this is what was fixed by the patch [06:29] doesn't work when I set it in control-center [06:29] alt + r just for testing and it works [06:29] hm, how do I specify in ccsm? [06:30] 1 ? or a different syntax? [06:30] oh, nevermind; it's a confusing UI, found it [06:30] pitti: you can ask for capturing your key combination [06:30] doesn't work there either [06:30] ? [06:30] * didrocks doesn't understand [06:30] it says 1 (and 2) [06:30] doesn't undertand at all [06:30] but it doesn't actually act [06:30] pitti: try anything without ctrll [06:31] ctrl is written as "primary", that doesn't seem right [06:31] tried 1 / 2, works [06:31] ah [06:31] can you try the same for switching windows in the backcgorund? [06:31] this also works in control-center [06:31] not using ctrl [06:32] yeah, I ensured the g-c-c integration worked [06:32] ooh [06:32] that works [06:32] nice, I have that key back! [06:32] so it's "just" control which is broken? [06:32] ok, so it's another issue [06:32] yeah, control seems to be mapped as "primary" [06:32] and no more "ctrl" [06:33] * pitti goes back to configuring his screen layout the normal way, having good IRC again [06:33] *happy* [06:33] I'm puzzled, nothing changed in ccsm, and no big changes in ibcompizconfig [06:33] new GTK? [06:33] yeah, possibly :/ [06:33] so, there is this bug [06:33] now, on the key tap not working [06:33] didrocks, gtk changed control to primary... [06:33] I would really like to reproduce them [06:33] DBO: oh? [06:33] so yeah, there is a mismatch somewhere now [06:34] ok, keybinding works - this primary that we should track [06:34] on the tapping now, you all have it? [06:34] * didrocks ties alt + left in weechat, chromium, always work, even the first time [06:35] RAOF: as you have the issue, can you do a build for me? [06:35] didrocks: Certainly. [06:35] Hah. *Yet another* person confused by [06:36] RAOF: lp:compiz, comment fix_925293.patch in debian/patches/series [06:36] pitti, didrocks: GTK (both 3 *and* 2) now map to . For some reason that I've not bothered to explore fully, except to bitch about. [06:37] RAOF: interesting, so compiz is still waiting for Ctrl, but ccsm, because of gtk is setting it to fo you :/ [06:39] didrocks: Yeah. Compiz (like apparently everything else which deals with keybindings) needs to deal with that. http://git.gnome.org/browse/gnome-control-center/commit/?id=7f5733b454e26c882d4e707b57cd03b104f57dd2 is how g-c-c handled it. [06:39] RAOF: oh, thanks for the pointer! [06:39] Or possibly ccsm. [06:39] yeah [06:39] Also, don't write a check like that :) [06:40] hum, there seem that there is a key control check [06:40] It doesn't like me to edit directly in ccsm [06:40] * didrocks tries in gconf [06:41] pitti: ok, Martin [06:41] pitti: in gconf-editor [06:41] pitti: you can edit your keybindings [06:41] for intance [06:42] 1 [06:42] that works [06:42] /apps/compiz-1/plugins//screen0/options [06:42] it's clearly due to ccsm + new gtk [06:43] phew, so the keybinding fix works, just need a little bit help of ccsm side [06:43] * didrocks started to think he will at the end revert all the patches we had [06:46] didrocks: hm, is that really a ccsm problem? I only touched that in g-c-c [06:46] didrocks: anyway, I'll keep Alt+B for now [06:46] pitti: yeah, I'm looking about this, it's weird [06:46] didrocks: in fact, that's what I had for a very long time until the menus grabbed them [06:47] but now the window handling keybinding seems to trump a menu with 'B', so that's fine for me ;) [06:50] ok [06:50] so g-c-c [06:50] still uses the keybinding integration [06:50] but store [06:50] in gconf [06:51] and show Control in the ui [06:51] at least, it all make sense [06:51] I think the easier way is in compiz making Primary == Control, otherwise, it's patching g-c-c, ccsm… [06:52] I still wonder if that is a Gtk bug [06:52] that seems like a rather pointless change [06:52] yeah, maybe better to wait for desrt on that one [06:53] RAOF: still building? :) [06:54] Yes [06:56] * RAOF heads out to the movies; I'll test once I get back. [07:02] ok, thanks RAOF :) [07:32] pitti, hi [07:33] hey tkamppeter [07:34] pitti, perhaps you know that Dwe have agreed on a printer driver packaging policy with Debian, where one part is to name the packages printer-driver-... [07:34] tkamppeter: yes [07:34] tkamppeter: and it seems we have most of that already [07:34] I updated the seeds along [07:35] pitti, due to this, all the printer driver packages are renamed now and the ubuntu-meta seeds are pulling the transitional packages. [07:35] pitti, there are two possibilities to correct this: [07:35] tkamppeter: https://launchpad.net/ubuntu/+source/ubuntu-meta/+changelog :) [07:35] tkamppeter: I might have missed some, but most are updated [07:35] c2esp already got renamed easlier [07:35] in 1.258 [07:37] pitti, fine, but perhaps we can go a more Debian-conforming way to include Debian's meta package printing-metas (source)/printer--drivers-all(binary). This is a meta package not installing any files, so no problem to introduce after FF. [07:37] tkamppeter: as long as it doesn't pull in a ton of extra packages, or universe ones, sure [07:43] pitti, it pulls printer-driver-escpr and c2050 from Universe. [07:44] pitti, we could MIR the two, use printing-metas with these two patched out, or not use printing-metas, WDYT? [07:46] tkamppeter: they are both tiny, so if they work and are well maintained, a MIR seems fine [07:47] pitti, is m2300w -> printer-driver-m2300w covered? I do not see this one. [07:47] we don't currently install this at all by default [07:47] pitti, printer-driver-escpr is well maintained and probably already widely used under Ubuntu because it is also available as OpenPrinting auto download. [07:47] supported-misc-servers: * m2300w # Driver for the Minolta magicolor 2300W/24000W color laser printers [07:48] I'll update that [07:48] done [07:49] pitti, does m2300w get installed with the standard installation, as all the others? [07:49] no [07:51] pitti, can we install it? The printers covered by m2300w should get equally treated as printers supported by other drivers. === philipballew is now known as philipballew_ [07:51] it's quite big (0.5 MB) === philipballew_ is now known as philipballew [07:51] so, as long as our CDs are horribly oversized, I'd veto any new major additions [07:52] tkamppeter: I wish we could install printer drivers on demand instead of having to ship them all by default [07:52] sounds like a nice UDS topic [07:52] we have a nice system using a PackageKit API for this now [07:52] so we could introduce this in an upstream friendly way [07:52] (i. e. not a pure jockey solution) [07:53] pitti, printing-metas would pull this one, too. [07:57] pitti, c2050 is for one very old Lexmark inkjet, discontinued ~10 years ago, driver for only 300 dpi color, probably made by a student to get his printer working and that's it. package continued to be maintained by Debian (only packaging and perhaps bug fixes) and auto-synced into Ubuntu ... [07:57] so that one doesn't seem very urgent then [07:58] pitti, I would only promote this to main to be able to sync printing-metas from Debian. I by myself even did not work on this package because the expected user community is extremely small. [08:03] Sweetshark: nice @ bug 900636; do you know whether the patches will be in the pending upload? [08:03] Launchpad bug 900636 in libreoffice "libreoffice ftbfs on armhf" [High,Fix committed] https://launchpad.net/bugs/900636 === david is now known as davidcalle [08:21] pitti, so how should we treat the printer driver seeding then? Simply as before, via ubuntu-meta? [08:38] tkamppeter: for now this works, so there's no urgency here [08:40] pitti, OK, and the next one will perhaps download all drivers ... [08:44] pitti, do you have any idea about bug 933973? Especially it seems to work under Debian but not under Ubuntu, though CUPS packages are identical under Debian and Ubuntu. [08:44] Launchpad bug 933973 in cups "Cups and related packages will not install without error in chroot" [Undecided,New] https://launchpad.net/bugs/933973 [08:56] hey [08:58] Greetings [08:59] Is there a way to reinstall all the ubuntu-desktop packages? [08:59] chrisccoulson, hey, did you see that t-bird is busted today? [08:59] I broke something [08:59] :> [08:59] bonjour seb128 [08:59] tkamppeter: the main difference is init.d vs. upstart [08:59] hey pitti, how are you? [09:00] rickspencer3, seems it's not? or only if you activate non standard extensions? [09:00] seb128: quite fine indeed; was quite happy to see this morning that precise_probs.html was beer-y and everything seems to be in good shape :) [09:00] seb128, well, I don't have non-standard extensions [09:00] Hi seb128 [09:00] it opens for me, but doesn't load any mailboxes or anything [09:01] rickspencer3, ok, dunno then, works for me ;-) dholbach and micahg just discussed known problem on #ubuntu-devel but that's due to extensions [09:01] hey davidcalle [09:01] pitti, great! [09:01] tkamppeter: there are no logs there, but I suppose it falls over when upstart is not running in the chroot; not sure whether jhunt knows a general solution to this [09:03] seb128, chrisccoulson I can still watch Taylor Swift videos, so we'll be okay for now [09:03] rickspencer3, it's only 9pm in the uk, wait for chrisccoulson to get coffee ;-) [09:03] hi! [09:03] pitti, can you assign the bug to the right upstart-in-chroot expert? Thanks. [09:03] lol [09:04] hey chrisccoulson, I'm logging a bug now [09:04] the bug is a missing file in the package. it would have been trivial for somebody to fix before i signed on this morning, rather than spending efforts discussing workarounds :) [09:04] seb128, I would need a very tiny amount of cd space, for the video lens + scope. [09:05] chrisccoulson, oh, someone discussed workaround? [09:05] chrisccoulson, what should I do? [09:05] just wait? [09:06] rickspencer3, you can start it in safe mode (thunderbird -safe-mode) [09:06] * rickspencer3 cancels bug report [09:07] tkamppeter: it could also be that it fails in post-start, and it needs to be more resilient to cupsd not running [09:07] chrisccoulson, thanks [09:08] as always, you demonstrated awesomeness all around [09:08] :) [09:08] hey chrisccoulson [09:08] hey pitti [09:08] how are you? [09:08] hey chrisccoulson [09:09] chrisccoulson: quite fine, thanks! [09:09] hi seb128 [09:20] pitti, btw the compiz merge request you reopen, the update is sitting in the oneiric sru queue for a week so technically it's merged ;-) [09:20] (I was just waiting on the autoimporter, I didn't bother actually using the vcs) [09:20] pitti, I'm closing it again [09:20] seb128: oh, I see; sorry [09:21] need to find some time to process SRUs [09:22] pitti, no worry [09:27] pitti - hmmm, any chance you could kill the thunderbird builds? [09:27] it's not actually enough to automatically recover from the problem that people are seeing [09:28] chrisccoulson: I can't [09:29] chrisccoulson: best to ask in teh IS channel [09:29] there is no interface for that aside from ssh'ing in as root and rm -r'ing the build tree, then let it die [09:29] hi chrisccoulson, is thunderbird issue you are discussing related to all panels turning blank with no content? [09:30] as in this - http://ubuntuone.com/2InpyFE2dUNV11zWijd7Ie ? [09:31] rye: bug 933951 [09:31] Launchpad bug 933951 in thunderbird "Messaging menu extension interacts badly with lightning in precise making thunderbird unusable" [High,Fix released] https://launchpad.net/bugs/933951 [09:31] not fixed, but workaround in there [09:32] it's trivially fixable. it would have been better to fix it earlier rather than talk about workarounds and assign it to me when i'm still asleep ;) [09:32] micahg, thank you! could not find the bug in lp, sorry [09:32] chrisccoulson: I had no idea what was causing it, if I did, I would've :) [09:42] pitti, thanks for the sru round! [09:42] you're welcome [09:46] I figured how to reinstall ubuntu desktop [10:01] grrrrr [10:01] launchpad! [10:03] oops an error, wasted my bug description! [10:05] seb128, CTRL-A, CTRL-C before clicking submit ... I have this as a conditional reflex. :) [10:05] yeah [10:05] seb128: go back [10:05] in most of the cases going back a page keeps the contents [10:05] firefox is pretty good at that [10:05] pitti, not with launchpad ajax stuff [10:05] I got my title back [10:05] oh [10:06] but not the description :-( [10:06] I was quite surprised that when appending a bug comment, the pending input text is still there even after closing and reopening firefox [10:15] pitti: bug 900636 patches will be in the next upload (although I think I will do a ppa upload without them first) [10:15] Launchpad bug 900636 in libreoffice "libreoffice ftbfs on armhf" [High,Fix committed] https://launchpad.net/bugs/900636 [10:15] Sweetshark: cool, thanks [10:15] Sweetshark: when do you plan to do this, BTW? we need to coordinate a bit with beta-1 [10:16] and armel takes very long these days [10:16] although it should be faster now, the babbages are gone [10:16] ppa upload today [10:18] main upload, hmm, I would have prefered to put it on a clean debian version, but we could drop that and just go with the ppa-version plus armhf fix [10:18] PPA version is building out of the packaging git? [10:19] Sweetshark: I'm asking because this currently fails all the automatic upgrade tests, so it hides the other upgrade issues we need to fix [10:20] where does it fail the automatic update tests? is that the unopkg issue again? === debfx_ is now known as debfx [10:25] right [10:25] also, it's one of the major blockers for moving armhf to our default arm flavour and dropping armel [10:26] Sweetshark: sorry that you have the most important package in the archive :) [10:29] pitti: stop flattering me ;) [10:29] * pitti sends a big ♥ to Sweetshark [10:30] * Sweetshark hugs pitti [10:30] Sweetshark: did you see http://www.youtube.com/watch?feature=player_embedded&v=lfcGEQRyaqI ? [10:31]