| Trevinho | mdeslaur: I've used another approach here https://code.launchpad.net/~3v1n0/unity/lockscreen-keys-disable/+merge/217528 | 00:25 |
|---|---|---|
| sarnold | Trevinho: that looks to me like it modifies some of the code you added recently to ensure it re-starts locked if it dies while locked -- does it still lock in that case? | 00:34 |
| Trevinho | sarnold: sure | 00:34 |
| Trevinho | sarnold: it just gets called in another cb function | 00:35 |
| Trevinho | sarnold: I could keep the old one, but duplicating the callbacks wasn't the nicest thing | 00:35 |
| sarnold | Trevinho: excellent :) thanks for reassuring me :) | 00:35 |
| bschaefer | looks good, and is way better then my branch | 00:35 |
| * bschaefer goes to reject his | 00:35 | |
| mdeslaur | Trevinho: thanks! | 01:13 |
| Trevinho | mdeslaur: if there are not other issues, I'm going to bed... That branch should be enough | 01:28 |
| Mirv | morning | 04:04 |
| Laney | hey! | 08:03 |
| seb128 | good morning desktopers | 08:11 |
| Laney | lut seb128 | 08:14 |
| seb128 | Laney, howdy, how are you? | 08:14 |
| * seb128 shakes fist at xnox | 08:15 | |
| Laney | very good thanks! and you? | 08:15 |
| seb128 | xnox, could you please stop uploading without considering the packaging vcs-es? | 08:15 |
| seb128 | Laney, I'm good thanks ;-) | 08:15 |
| Laney | naughty xnox | 08:16 |
| seb128 | especially that in this case he stepped over a version already used for a SRU | 08:16 |
| seb128 | which made the SRU be rejected after a week of waiting in the queue | 08:16 |
| seb128 | *great* | 08:16 |
| mlankhorst | so touch can corrupt xorg-server :P | 08:23 |
| mlankhorst | anyone here with a touchscreen and some interest in valgrind? | 08:23 |
| seb128 | well, no special "interest" but I can probably help getting debug info | 08:24 |
| mlankhorst | yeah just spawning xorg-server + a unity session and reproducing bug 1311828 and bug 1298727 is enough. | 08:28 |
| ubot2 | Launchpad bug 1311828 in xorg-server (Ubuntu) "Xorg crashed with SIGABRT" [Medium,New] https://launchpad.net/bugs/1311828 | 08:28 |
| ubot2 | Launchpad bug 1298727 in xorg-server (Ubuntu) "Xorg crashed with SIGABRT in point_on_screen()" [Medium,Incomplete] https://launchpad.net/bugs/1298727 | 08:28 |
| mlankhorst | something like "/usr/bin/valgrind --keep-stacktraces=alloc-and-free --show-reachable=yes --track-fds=yes --leak-check=full --error-limit=no --freelist-vol=50000000 --freelist-big-blocks=10000 --track-origins=yes --leak-resolution=high --malloc-fill=ef --free-fill=df /usr/bin/Xorg "$@" -core -verbose 10" | 08:28 |
| mlankhorst | and install xserver.*dbg$ | 08:29 |
| seb128 | k, I'm trying to do that in a bit, need to update my test config first, install valgrind, dbgs, etc and see if I can reproduce the bug | 08:35 |
| seb128 | where do you put that command? do you replace the X binary by that or..? | 08:35 |
| mlankhorst | that's one way to do it :P | 08:36 |
| mlankhorst | I created an /etc/X11/X2 shellscript, and symlinked /etc/X11/X to it (that file must be a symlink or X won't start) | 08:36 |
| mlankhorst | http://paste.debian.net/96346/ my /etc/X22/X2 | 08:37 |
| seb128 | thanks | 08:38 |
| Laney | x22! | 08:38 |
| mlankhorst | oops :P | 08:38 |
| mlankhorst | oh and probably the dbg files for libdrm.*dbg$ and libpixman.*dbg$ | 08:40 |
| xnox | seb128: argh, sorry. which package did i step over like that? | 08:46 |
| seb128 | xnox, empathy | 08:46 |
| xnox | seb128: well, use SRU version numbers for SRUs? =))))) | 08:47 |
| seb128 | xnox, what is a SRU version? | 08:48 |
| xnox | 3.8.6-0ubuntu9.1 | 08:48 |
| xnox | as per SRU docs... | 08:49 |
| seb128 | xnox, that wouldn't have changed the fact that the vcs had work to be uploaded that should have been included in that upload, instead of doing a no change upload | 08:51 |
| xnox | yeah, sorry. | 08:51 |
| seb128 | no worry, please be careful with desktop packages next time though, most have packaging vcs-es | 08:52 |
| xnox | some are out of date, some are not. | 08:52 |
| xnox | i want a way to do version numbers for rebuilds that are not reusing any other upload numbers ideally. to avoid contention. | 08:53 |
| ricotz | seb128, hi | 09:09 |
| ricotz | just a short question ;) | 09:09 |
| ricotz | will utopic target gnome 3.12 or are you going straight for 3.13/14? | 09:09 |
| seb128 | lol | 09:09 |
| seb128 | ricotz, hey | 09:09 |
| ricotz | i guess the conservative approach turned out fine for you | 09:10 |
| seb128 | ricotz, we plan to discuss the topic during the meeting later today, but I can tell you for sure not going to 3.13 | 09:10 |
| seb128 | I would rather ponder the "stay on 3.10" | 09:10 |
| seb128 | but realistically I think we are going to end up doing 3.12 | 09:10 |
| ricotz | alright ;) | 09:10 |
| seb128 | and yes, the "stay on stable" serves us pretty well | 09:11 |
| ricotz | 3.12 sounds fine since there are a lot synced already too | 09:11 |
| seb128 | we didn't have too much fire fighting since we do that and quality went up | 09:11 |
| ricotz | right ;) | 09:12 |
| seb128 | well, 3.12 synced, so things depwait on a newer gtk? | 09:12 |
| seb128 | I don't see gtk 3.12 being done/ready before some time | 09:12 |
| ricotz | no, just some minor components which doesnt require gtk 3.12 | 09:12 |
| seb128 | k, that's fine then | 09:12 |
| ricotz | gtk 3.12 is in gnome3-staging so testing it is possible | 09:13 |
| seb128 | is that a proper update? | 09:13 |
| Laney | with patches? | 09:13 |
| ricotz | yes | 09:13 |
| seb128 | or another of those updates dropping the patches that were too difficult to port? | 09:13 |
| ricotz | please just give it a short look and judge about it | 09:14 |
| ricotz | thanks, bbl | 09:15 |
| seb128 | is there any known issue with themes or overlay scrollbar? | 09:16 |
| Laney | does seem to have all the patches, only change is git backports | 09:16 |
| seb128 | yeah | 09:16 |
| seb128 | I just compared the series | 09:16 |
| Laney | and the other ones didn't require much updating ... | 09:17 |
| Laney | what is this wizardry | 09:17 |
| seb128 | larsu, ^ maybe you want to have a look/try gtk 3.12 from https://launchpad.net/~gnome3-team/+archive/gnome3-staging/+packages?field.name_filter=gtk%2B3 ? | 09:17 |
| Laney | is this like bearing the fruits of upstreaming stuff or something :P | 09:17 |
| seb128 | there must be some runtime issues with overlay-scrollbar or something | 09:18 |
| seb128 | one gotcha is going to be the text-wrapping/dialogs I think | 09:20 |
| seb128 | but we can distro patch that out before release if needed | 09:20 |
| Laney | yep, debian will probably do that too | 09:20 |
| seb128 | looking to the NEWS entries, that seems not a crazy cycle | 09:21 |
| seb128 | that's good news | 09:21 |
| seb128 | I'm going to try it before the meeting today | 09:22 |
| seb128 | somebody has been browsing e.u.c for lightdm issue and clicking the "create" button | 09:31 |
| * mlankhorst pokes seb128 | 09:34 | |
| seb128 | yes? | 09:34 |
| seb128 | I'm starting the test box for your issue | 09:35 |
| seb128 | sorry, got busy until now with other things | 09:35 |
| mlankhorst | k | 09:46 |
| mlankhorst | np :) | 09:46 |
| === davidcalle_ is now known as davidcalle | ||
| darkxst | hey seb128, Laney | 09:49 |
| seb128 | hey darkxst | 09:49 |
| darkxst | seb128, really? staying on 3.10? as if that well happen ;) | 09:53 |
| seb128 | easiest way to avoid gtkheaderbars... | 09:54 |
| larsu | ricotz: where do you keep the packaging branch for gtk? | 09:55 |
| darkxst | seb128, did the entire conversation last night, just fly past you? | 09:55 |
| darkxst | larsu, we don't have any packaging branches currently, going to look at setting that up this cycle | 09:55 |
| larsu | darkxst: ah okay, thanks | 09:55 |
| seb128 | darkxst, which one? | 09:56 |
| seb128 | on how to handle gtkheaderbars? | 09:56 |
| darkxst | seb128, yes, gtkheaderbars! | 09:56 |
| seb128 | or on the conflict of interest between flavors? | 09:56 |
| seb128 | but, no, none did fly past no | 09:57 |
| darkxst | seb128, and ignoring the WN hints to get back titlebars | 09:57 |
| darkxst | WM even | 09:57 |
| seb128 | I'm still not convinced we can make gtkheaderbar a non-regression for !gnome-shell users | 09:57 |
| seb128 | I'm not convinced that wm bar + headerbar without control is not a regression in UI over what we have | 09:57 |
| seb128 | I need to test on real cases to see how it looks/feel | 09:58 |
| darkxst | seb128, without window controls and with titlebars its not much difference to what in archives currently | 09:58 |
| seb128 | like there is no title in the headerbar? | 09:58 |
| darkxst | seb128, huh? you guys would have a titlebar with a *title*! | 09:59 |
| larsu | seb128: I want to talk about this at the hackfest tomorrow | 09:59 |
| larsu | there are two approaches for us | 09:59 |
| larsu | (1) go csd, but with our theme | 09:59 |
| larsu | (2) make headerbars into primary toolbars by adding a titlebar | 09:59 |
| darkxst | larsu, upstream won't take 2, so that would have to be a distro patch | 10:00 |
| seb128 | what do you mean with (2)? | 10:00 |
| seb128 | the title issue is | 10:00 |
| seb128 | what happens to http://curiousdtu.files.wordpress.com/2014/02/gedit2.png?w=774 if we add a wm bar | 10:00 |
| seb128 | do we have "unsaved document" 3 times ? | 10:00 |
| seb128 | one in the wm bar, one in the headerbar bellow it, one in the tab? | 10:00 |
| larsu | darkxst: I'm not so sure about that | 10:01 |
| larsu | seb128: no, it would hide the title in the toolbar | 10:02 |
| seb128 | larsu, how would it look for apps that have only a title and a close button if we hide both | 10:02 |
| seb128 | would we get an empty headerbar? | 10:02 |
| seb128 | or would it get hidden? | 10:02 |
| larsu | I don't know about that case | 10:03 |
| larsu | most of the time, apps have additional buttons there | 10:03 |
| darkxst | larsu, when I discussed with them, they seemed to think ignoring the WM hints in Unity was the way to go | 10:03 |
| larsu | like in the screenshot you posted | 10:03 |
| seb128 | larsu, https://lh6.googleusercontent.com/-Ixfpn57HBL8/UtbDAz_crhI/AAAAAAAAAaM/xwvAaPy8Ebc/w480-h500-k/gnome%2Bcalculator%2B3.11.3%2Bfedora%2B21%2Brawhide.png | 10:03 |
| desrt | seb128: i think i've seen this link somewhere before ;) | 10:03 |
| seb128 | desrt, no, different one | 10:03 |
| seb128 | that's not a buggy case this time ;-) | 10:04 |
| desrt | oh. why do you show it, then? :) | 10:04 |
| larsu | darkxst: who is "them"? | 10:04 |
| seb128 | just an example of headerbar with only a title and button | 10:04 |
| seb128 | desrt, ^ | 10:04 |
| seb128 | desrt, because of <seb128> larsu, how would it look for apps that have only a title and a close button if we hide both | 10:04 |
| darkxst | larsu, mclassen and co | 10:04 |
| larsu | desrt is meaning to talk to him | 10:04 |
| desrt | was gonna yesterday but got busy with menus | 10:04 |
| larsu | seb128: fair point. Maybe we should hide it then | 10:04 |
| larsu | seb128: I'm not sure (2) is the better option, and in fact desrt just mentioned some compelling reasons for (1) | 10:05 |
| larsu | seb128: I'm still pondering and haven't thought about all of the issues | 10:05 |
| seb128 | I'm unsure what (1) mean | 10:05 |
| seb128 | is that doing what GNOME is doing? | 10:05 |
| larsu | yes, but changing it to look more like unity | 10:05 |
| seb128 | e.g hidding the wm controls and making the theme look "native enoguh" | 10:06 |
| seb128 | right | 10:06 |
| larsu | dark theming, different buttons etc | 10:06 |
| larsu | ya | 10:06 |
| seb128 | my concern that we would never have perfect consistency that way | 10:06 |
| seb128 | but I'm happy to be proved wrong | 10:06 |
| desrt | seb128: we decide to have a desktop based on five toolkits.... | 10:06 |
| desrt | we will never have perfect consistency, ever | 10:06 |
| larsu | I'm not sure "perfect consistency" is what we should optimize for | 10:06 |
| larsu | many apps are going that route | 10:06 |
| larsu | for example chromium | 10:06 |
| seb128 | chromium is much better integrated than gtkheaderbars | 10:07 |
| seb128 | it even has a setting to use the name wm controls | 10:07 |
| darkxst | and QT apps pretending to be GTK ones like USC! | 10:07 |
| seb128 | name->native | 10:07 |
| seb128 | well, one toolkit you can dream about | 10:07 |
| seb128 | but it's not going to happen | 10:07 |
| seb128 | unless you want to replace skype, eclipse, libreoffice, firefox, chrome | 10:07 |
| seb128 | and I don't see GNOME achieving that either | 10:08 |
| darkxst | seb128, epiphany is pretty slick these days ;) | 10:08 |
| seb128 | so reality is that we have to deal with apps using different toolkits | 10:08 |
| seb128 | darkxst, if you don't care about security in your browser yes | 10:08 |
| seb128 | webkitgtk doesn't do security updates | 10:08 |
| desrt | epiphany is great -- for webapps, with trusted sites :) | 10:09 |
| darkxst | seb128, yes I know... that is why we still ship firefox ;( | 10:09 |
| seb128 | which is why no distro ships with epiphany as a webbrowser | 10:09 |
| desrt | using it as a general-purpose browser is pretty much insane | 10:09 |
| desrt | (and unusable, honestly) | 10:09 |
| Sweetshark_gc | I hear there is a volunteer to finish the gtk3 native port of libreoffice in the channel? | 10:09 |
| seb128 | right | 10:09 |
| seb128 | Sweetshark_gc, would that be you? ;-) | 10:09 |
| Sweetshark_gc | seb128: *cough* no? | 10:09 |
| seb128 | Sweetshark_gc, start by fixing those menus! | 10:10 |
| Sweetshark_gc | *grumble* | 10:10 |
| darkxst | Sweetshark, and add gtkheaderbars ;) just to annoy seb128 ;) | 10:10 |
| seb128 | larsu, but yeah "perfect consistency" might not be the goal, I would still prefer to avoid user-experience-regressions in our default apps if possible | 10:10 |
| seb128 | larsu, but solutions to that might be to patch like we did for menus | 10:10 |
| chrisccoulson | seb128, I'd be open to someone contributing a gtk port of oxide if you want a webengine that's maintained by the security team ;) | 10:10 |
| seb128 | larsu, or to replace those apps by some "written for traditional desktops" | 10:11 |
| seb128 | larsu, e.g we might decide that e.g eog design goals are too far from ours and that we would be better served by another image viewer | 10:11 |
| desrt | seb128: taking a step back, headerbars are an awesome step forward | 10:11 |
| seb128 | desrt, depending for who, not for app writers that want to target the different desktops | 10:12 |
| desrt | seb128: i said to take a step back :p | 10:12 |
| seb128 | they give their users a pretty suboptimal experience | 10:12 |
| desrt | i agree that they're not a great fit if you just drop one app into a desktop like this | 10:12 |
| desrt | but what i'm trying to say is, honestly, i think we should try to move more toward such designs ourselves | 10:13 |
| desrt | the issue here is only the consistency one | 10:13 |
| desrt | (and i agree that is an issue) | 10:13 |
| seb128 | right, I agree with that | 10:13 |
| * darkxst hands seb128 a fork ;) | 10:17 | |
| * desrt hands seb128 a knife and a steak | 10:17 | |
| * larsu hands seb128 some red wine | 10:18 | |
| larsu | seb128: we'd have user experience regressions too if we switch apps, no? | 10:19 |
| seb128 | larsu, not especially ? if some app is better ? | 10:24 |
| seb128 | larsu, like we replaced f-spot by shotwell, I don't consider that an user regression | 10:24 |
| seb128 | it's different, it doesn't mean less integrated/less easy to use/less features | 10:25 |
| desrt | seb128: i think the trouble is in finding high quality apps :) | 10:25 |
| desrt | for better or worse, gnome (still) has some of the best stuff around | 10:25 |
| larsu | seb128: I meant the confusion that comes with changing the app. But yeah, if it's much better that weighs more | 10:25 |
| seb128 | desrt, well, one possible one could be evince -> okular for example (note that I didn't look at okular, but poppler upstream keep using it as an example of better pdf viewed than evince) | 10:26 |
| desrt | seb128: poppler upstream are kde developers :p | 10:26 |
| seb128 | I know, okular seems to be a good pdf viewer though | 10:26 |
| seb128 | anyway no point to discuss specific examples | 10:26 |
| desrt | specific examples are good to discuss, i think | 10:27 |
| seb128 | I'm also sure we would have no difficulty finding a decent calculator | 10:27 |
| desrt | okular might be good | 10:27 |
| seb128 | it's not like gnome-calculator was the perfect calc | 10:27 |
| desrt | but if we want to switch away from nautilus or something? | 10:27 |
| seb128 | we would take nemo | 10:27 |
| desrt | scary :) | 10:27 |
| seb128 | that would give us back split view | 10:27 |
| desrt | why not caja? | 10:27 |
| seb128 | and other stuff our users complain about loosing | 10:27 |
| ochosi | sorry to chip in, but are you discussing moving towards headerbars? | 10:27 |
| * larsu missed those games the last few cycles | 10:27 | |
| seb128 | I don't know what caja is | 10:27 |
| larsu | ochosi: yes | 10:27 |
| desrt | mate-nautilus | 10:27 |
| darkxst | ipython is a pretty good calculator ;) although probably not for thte masses ;) | 10:28 |
| ochosi | from xubuntu POV (if that is of any concern/interest to you guys) we were rather happy about you patching out the headerbars in 14.04 | 10:28 |
| ochosi | s/xubuntu/xubuntu's/ | 10:28 |
| larsu | yeah... the problem is that that will become increasingly hard | 10:29 |
| darkxst | ochosi, only nautilus and epiphany were patched, everything else was just avoided | 10:29 |
| desrt | also: headerbars, in their own right, are good | 10:29 |
| larsu | and seb128 is afraid that we'll get another patch madness like we have with menus now | 10:29 |
| ochosi | right, that's understandable. but personally i think it'd be better to talk to upstream about providing more options for !gnome-shell desktops | 10:30 |
| ochosi | just "dealing with what we got" is what xubuntu alone is forced to do, we're a tiny team and we have no leverage/voice | 10:30 |
| desrt | seb128: i hate to appeal to the media, but it seems that all i hear about anymore is gnome is getting better UI and unity is getting worse... | 10:30 |
| darkxst | ochosi, upstream did quite some work with for example allowing WM button layouts within the headerbars | 10:31 |
| desrt | i think we should not be afraid to follow gnome on some of the improvements | 10:31 |
| larsu | ochosi: talking to them about more options is asking them to do the work, which is a bit unreasonable | 10:31 |
| seb128 | desrt, not sure what you call "media", I didn't read anything about GNOME vs Unity for a long time and all the trusty reviews I read stated that Unity was "a polished desktop now" | 10:31 |
| darkxst | but they are not going to make gtkheaderbars optional, and atleast in my discussions with them, they seem to think the WM should just ignore the MWN hints for decorations/titlebars | 10:31 |
| desrt | seb128: i just searched 'ubuntu unity' and the first item that came in google news is "Ubuntu 14.04 review: Missing the boat on big changes | 10:32 |
| seb128 | desrt, is media = reddit ? | 10:32 |
| desrt | While a new kernel should mean better performance, Canonical's UI troubles persist. " | 10:32 |
| desrt | arstechnica in this case | 10:32 |
| ochosi | larsu: whatever happened to the good practice of a bit of backward compatability? anyway, i understand, but i think that'd be a good thing even if patches would have to be contributed. maybe i'm too optimistic about "making a voice heard" | 10:32 |
| ochosi | seb128: +1, i haven't read much about that either | 10:33 |
| larsu | ochosi: voices are heard, it's just that noone has time for it | 10:33 |
| ochosi | darkxst: yeah, but being able to disable headerbars would be more desirable. there's a lovely post by martin grässlin about the shortcomings of headerbars (not being able to kill a hanging app,...) | 10:33 |
| seb128 | desrt, yeah for crappy titles | 10:34 |
| seb128 | "What's missing | 10:34 |
| seb128 | Mir and Unity 8 did not make the cut, but they will be coming eventually (14.10 looks pretty likely to see at least xMir enabled by default)" | 10:34 |
| desrt | media on new gnome release seems to be "hey... this is finally getting really nice... too bad nobody ships it" | 10:34 |
| seb128 | desrt, so arstechnica is speaking about the LTS not having the newest kernel, Mir or unity8 | 10:34 |
| desrt | seb128: right | 10:34 |
| seb128 | but the unity part of the review is rather positive | 10:34 |
| desrt | in general i see the biggest thing said about this release is "kinda boring... not much changes..." | 10:34 |
| seb128 | right, that's a LTS | 10:35 |
| desrt | not much changes since last lts :p | 10:35 |
| seb128 | desrt, that review you mentioned states "Ubuntu is one of the most polished desktops around, certainly the most polished in the Linux world, " | 10:36 |
| seb128 | in the conclusion | 10:36 |
| desrt | seb128: and the very same article, in the conclusion, talks about the mess that all of our patching has caused us :p | 10:36 |
| seb128 | they have more issues with things like nautilus dropping features, and they say it's coming from GNOME | 10:36 |
| seb128 | "what does it say that it still can't make menus behave consistently?" | 10:37 |
| seb128 | haha | 10:37 |
| seb128 | desrt, well, I don't have the same reading | 10:37 |
| seb128 | to me their issues is what I was pointing before | 10:37 |
| seb128 | design changes from GNOME | 10:37 |
| darkxst | seb128, gnome-shell has improved in leaps and bounds since last LTS | 10:37 |
| seb128 | that are making apps no consistent | 10:37 |
| ochosi | +1 | 10:37 |
| desrt | seb128: seems like the biggest complaint is with global menu, in fact | 10:37 |
| desrt | but i think maybe we should stop arguing over what is written in the press | 10:38 |
| seb128 | you are the one who started it! | 10:39 |
| seb128 | but agreed | 10:39 |
| desrt | ya | 10:39 |
| desrt | even when i started it i knew it was a bad idea :p | 10:39 |
| seb128 | I think your view of the press is biased as well | 10:39 |
| desrt | maybe | 10:39 |
| desrt | but one thing that is pretty clear is that people are liking the direction of gnome | 10:39 |
| desrt | so if they don't like how gnome apps are integrating in ubuntu then this is either an issue of integration or an issue of the patching that we've done | 10:40 |
| seb128 | desrt, I'm unsure I agree with that, some people like where GNOME is going, some hate it | 10:41 |
| seb128 | same for Unity | 10:41 |
| darkxst | desrt, or an issue of some gnome apps being very old (think gnome-terminal) | 10:41 |
| desrt | seb128: i think people are upset over some specific things in gnome -- i'm one of them | 10:41 |
| desrt | but the overall design idea appears to be paying off | 10:41 |
| seb128 | I read lot of comments from user who like mint/cinnamon/elementary/xfce better than GNOME3 or Unity | 10:42 |
| desrt | ya... of course | 10:42 |
| desrt | i hear that some crazy people even like kde... | 10:42 |
| seb128 | lol | 10:42 |
| seb128 | I don't think GNOME is winning the "battle of minds" | 10:42 |
| desrt | i'm just saying that if we want to continue to have gnome apps, we may want to adopt some of their new approaches | 10:42 |
| seb128 | we have more fragmention than before in anything | 10:42 |
| desrt | they're not bad | 10:43 |
| seb128 | and I don't see a clear winner | 10:43 |
| seb128 | in->if | 10:43 |
| desrt | and our increasingly complicated attempts to remain in the past are starting to hurt us more and more | 10:43 |
| seb128 | well | 10:43 |
| desrt | (the past = make gnome apps look like they used to) | 10:43 |
| darkxst | desrt, well particularly hurting Ubuntu GNOME! | 10:43 |
| seb128 | unity7/our current desktop is meant to be a "stable developer desktop" | 10:43 |
| seb128 | our innovations go to unity8 | 10:44 |
| * desrt kinda wishes gnome had this kind of focus... | 10:44 | |
| seb128 | that has its own set of apps, etc | 10:44 |
| * darkxst can't believe that some people prefer to crawl through applications menus, rather types a few letters to search for said app | 10:44 | |
| seb128 | so all the problems we discuss are going to be resolved there | 10:44 |
| seb128 | the friction is on the "traditional desktop" | 10:44 |
| darkxst | seb128, how so? unity8 will just further disjoint the "traditional desktop" people? | 10:45 |
| desrt | darkxst: most people always had some sort of launcher icons for their favourites | 10:45 |
| seb128 | darkxst, what do you mean? | 10:46 |
| darkxst | desrt, that is still there in gnome-shell and unity (I presume) | 10:46 |
| seb128 | darkxst, well, unity8 comes with new apps using the new toolkit, so it's going to resolve the conflict of (ab)using GNOME apps to make them match our design | 10:46 |
| darkxst | seb128, touch apps on desktop ;) | 10:47 |
| desrt | fwiw, i don't care too much about ubuntu gnome... | 10:48 |
| larsu | darkxst: the idea is that those will be adjusted for desktop usage and more laptops will have touchscreens | 10:48 |
| desrt | my primary concern is the increasing amount of developer resource we spend on patching things... | 10:48 |
| larsu | the question is what we do in the meantime | 10:48 |
| larsu | which might be a couple of cycles | 10:48 |
| seb128 | right | 10:48 |
| seb128 | well, as said we basically have 2 desktops | 10:49 |
| desrt | could maybe stop updating all gnome apps | 10:49 |
| desrt | this is a pretty viable idea, really... | 10:49 |
| seb128 | unity7 is a "traditional" developer desktop, I don't see much design changes or innovation going there | 10:49 |
| seb128 | that's what I was sort of suggesting | 10:49 |
| desrt | since most of the changes in gnome these days are changing to new UIs.... and then we just want to reverse those anyway | 10:49 |
| seb128 | just keep unity7 as a solid base, just fix issues | 10:49 |
| desrt | kinda pointless | 10:49 |
| desrt | imho this is not the right approach, but i certainly like it better than moving to new versions and then hacking their UIs halfway back to the old way | 10:50 |
| darkxst | desrt, the entire point of Ubuntu GNOME is building a community around gnome desktop and make it work well in Ubuntu | 10:51 |
| desrt | darkxst: i know | 10:51 |
| darkxst | suggesting to hold everything back will just kill our project | 10:51 |
| desrt | but i guess it's not really the concern of canonical... | 10:51 |
| darkxst | and the potential developers that come from it ;) | 10:52 |
| desrt | darkxst: you've elected to fight an uphill battle... | 10:52 |
| desrt | unless we want to start forking everything (unity-nautilus, unity-evince, etc.) like we did g-s-d and g-c-c, you're faced with what is effectively a fork of gnome, trying to make it work like upstream | 10:54 |
| darkxst | desrt, I am not going to fork upstream projects! Canonical should be forking the oldies they want to keep | 10:55 |
| desrt | ya... maybe there is some good argument for this | 10:56 |
| larsu | seb128: 3.12 is working fairly nicely | 10:56 |
| desrt | we did it with g-s-d and g-c-c, after all | 10:56 |
| larsu | seb128: first issues I'm seeing are titlebar-less dialogs | 10:56 |
| seb128 | larsu, good news! | 10:56 |
| desrt | but then you get the burden of packaging the gnome releases | 10:56 |
| larsu | seb128: even o-s seems to work | 10:56 |
| seb128 | \o/ | 10:56 |
| larsu | I haven't done extensive testing yet, though | 10:56 |
| desrt | this makes me very happy | 10:56 |
| desrt | it would have been extremely bad news if we would have had to get rid of o-s | 10:56 |
| seb128 | desrt, if we are going down to "fork" we can as well share the fork, and use e.g nemo... | 10:56 |
| desrt | or caja | 10:57 |
| desrt | take a look at mate, seriously... | 10:57 |
| desrt | it's basically like releasing old versions of gnome :p | 10:57 |
| seb128 | heh | 10:57 |
| desrt | just keep updating the libraries -- this is all most people care about | 10:58 |
| desrt | "developers", after all | 10:58 |
| larsu | desrt: it's what you care about | 10:58 |
| === MacSlow is now known as MacSlow|lunch | ||
| desrt | larsu: in so far as i care about people like yorba who care about things like which libraries are available on ubuntu | 10:59 |
| larsu | desrt: they also care about what kind of ui and platform integration we have | 11:00 |
| larsu | and it's a pain to support more than one for them | 11:00 |
| desrt | larsu: ya... but we can keep this looking OK by staying with old versions | 11:00 |
| larsu | man, why are we talking on irc? | 11:00 |
| larsu | I can even see your irc window from here | 11:00 |
| desrt | because we're not red hat ;) | 11:00 |
| larsu | haha | 11:00 |
| seb128 | desrt, larsu: anyway, to focus back the discussion | 11:02 |
| seb128 | let's make GtkHeaderBar work as good as we can on !gnome-shell (Unity being the main focus) | 11:02 |
| seb128 | that's the first step | 11:02 |
| desrt | this sounds good | 11:02 |
| seb128 | once we think we have "as good as we can get", we can see how apps feel like | 11:03 |
| seb128 | and then decide on the next steps for those | 11:03 |
| seb128 | like if we feel gtkheaderbar are integrated enough | 11:03 |
| seb128 | or if we need to get out of our way to resolve extra issues | 11:03 |
| larsu | seb128: what do you mean by that? My suggestion (1) or (2) from earlier? | 11:03 |
| desrt | there are a few things that are needed to get it working right... but not majorly huge ones | 11:03 |
| larsu | (make them look native in unity or make them be toolbarS) | 11:03 |
| seb128 | larsu, I'm sure, whichever you feel like is going to give us the best experience on Unity | 11:04 |
| seb128 | *unsure* | 11:04 |
| desrt | sounds like we have some work for the coming days | 11:04 |
| larsu | ya, this is my main goal for the hackfest | 11:04 |
| seb128 | great | 11:05 |
| larsu | seb128: interesting. I'm unsure as well :) | 11:05 |
| seb128 | sorry for all the side discussions | 11:05 |
| larsu | no worries | 11:05 |
| alkisg | attente: hi, could I bother you a bit about http://bazaar.launchpad.net/~ubuntu-desktop/gnome-settings-daemon/ubuntu/revision/437.1.10, and in particular about these lines? | 11:10 |
| alkisg | 344+ if (n_sources < 2 || g_strcmp0 (g_getenv ("XDG_CURRENT_DESKTOP"), "Unity") == 0) | 11:10 |
| alkisg | 345 strip_xkb_option (options, "grp:"); | 11:10 |
| alkisg | These lines break keyboard layout switching in applications that grab the keyboard, like SDL applications, so we cannot type e.g. Greek in e.g. tuxpaint, tuxtype, teeworlds etc. | 11:10 |
| alkisg | See also e.g. my comment #7 in https://bugs.freedesktop.org/show_bug.cgi?id=42244 | 11:10 |
| ubot2 | Freedesktop bug 42244 in Server/Input/Core "Multimedia keys become unresponsive in full-screen applications" [Normal,Reopened] | 11:10 |
| alkisg | (it's not only about multimedia keys anymore, now it's significantly more important, all those applications become useless for non-latin environments) | 11:11 |
| attente | alkisg, we disabled it because layout switching is done differently in unity and we didn't want xkb switching to conflict with that | 11:13 |
| alkisg | attente: and, is there any provision for the issue I mentioned? | 11:13 |
| alkisg | Because I don't think you'll be able to allow keyboard switching without letting xkb manage it... | 11:14 |
| alkisg | (in apps that grab the keyboard) | 11:14 |
| seb128 | alkisg, no, I think it's one of those GNOME design decisions where you are going to need to wait for wayland/Mir to have it working back correctly | 11:14 |
| alkisg | seb128: in Fedora they have it working like I mentioned | 11:15 |
| alkisg | It's an ubuntu-specific patch | 11:15 |
| attente | alkisg, sorry, i'm reading the bug report, but not really sure how the grp: option is related | 11:15 |
| seb128 | alkisg, the freedesktop bug states "(It's the same in other distros too, it's not Ubuntu specific)" | 11:15 |
| alkisg | seb128: there are many many bugs about it, it's a lot more than this one, | 11:15 |
| seb128 | alkisg, also the xorg grabbing is nothing Ubuntu specific | 11:15 |
| alkisg | ...I'll probably need to describe it better | 11:15 |
| seb128 | alkisg, overstatement like that don't help, can you just focus on the issues? | 11:16 |
| alkisg | So, there are many issues wrt keyboard, switching etc, I'll ignore them for now and only focus on the issue I described above, | 11:16 |
| alkisg | i.e., how to switch layouts when unity-settings-daemon can't get an event because of the keyboard grab | 11:16 |
| seb128 | alkisg, GNOME stopped using xserver group switching in favor of grabbing in the shell and doing changes themself from there | 11:16 |
| seb128 | we do the same in Unity now | 11:16 |
| alkisg | seb128: I've been talking with upstream gnome for the past 3 days | 11:16 |
| seb128 | and? | 11:17 |
| alkisg | The issue I'm mentioning right now, is ubuntu-specific | 11:17 |
| alkisg | ...but, you are right in what you say, | 11:17 |
| seb128 | do you use Unity? | 11:17 |
| alkisg | and that causes a few other issues, like e.g. that when I use xkb to switch layouts, the indicators don't get updated | 11:17 |
| seb128 | or gnome fallback? | 11:17 |
| alkisg | I tried in Unity, Gnome-fallback, gnome-shell etc | 11:17 |
| seb128 | they all have the issue? | 11:17 |
| alkisg | The first two, but, if I launch unity-settings-daemon with a different XDG_CURRENT_DESKTOP, then fallback works two, and I imagine unity will too, I haven't tried that there | 11:18 |
| alkisg | So, I'm trying to see why that thing was added, in order to help patching it elsewhere | 11:18 |
| seb128 | hum | 11:19 |
| alkisg | That one: || g_strcmp0 (g_getenv ("XDG_CURRENT_DESKTOP"), "Unity") == 0) | 11:19 |
| seb128 | what you said doesn't make mutch sense | 11:19 |
| seb128 | the old way was to let g-s-d/u-s-d do the grabbing | 11:19 |
| alkisg | setxkbmap -query, says: options: grp:alt_shift_toggle,grp_led:scroll | 11:19 |
| seb128 | in Unity session compiz is grabbing the keys | 11:19 |
| seb128 | same as gnome-shell is doing | 11:19 |
| seb128 | so technically compiz/Unity sessions should behave the same as gnome-shell | 11:20 |
| alkisg | That's the correct thing to have. But, that part of the code, strips the alt_shift_toggle, and we can't switch locale in e.g. tuxtype anymore | 11:20 |
| alkisg | Yes, I don't mind that part | 11:20 |
| seb128 | or did they apply some fixes in newer version we didn't backport/do something different? | 11:20 |
| alkisg | It does cause issues, but other issues, not what I'm trying to describe | 11:20 |
| alkisg | (02:19:45 μμ) seb128: in Unity session compiz is grabbing the keys => I.e. I'm not asking to remove that, | 11:21 |
| alkisg | I'm only asking to allow xkb have a layout switching shortcut as well | 11:21 |
| alkisg | Same as gnome-shell does | 11:21 |
| seb128 | hum? | 11:21 |
| seb128 | the stripping is done by g-s-d | 11:21 |
| seb128 | https://git.gnome.org/browse/gnome-settings-daemon/commit/?id=fc3676f0457789307e49d7abbf8115457a25e479 | 11:21 |
| seb128 | same upstream | 11:21 |
| seb128 | is that commit your issue? | 11:22 |
| alkisg | No | 11:22 |
| alkisg | I can solve my issue just by running: XDG_CURRENT_DESKTOP=Gnome unity-settings-daemon --replace & | 11:22 |
| seb128 | well, that makes u-s-d do the grabbing again | 11:23 |
| alkisg | I.e. just by omitting the debian/patches/unity-modifier-media-keys.patch | 11:23 |
| seb128 | instead of using the compositor | 11:23 |
| alkisg | seb128: see that one: http://bazaar.launchpad.net/~ubuntu-desktop/gnome-settings-daemon/ubuntu/revision/437.1.10 | 11:23 |
| alkisg | 344+ if (n_sources < 2 || g_strcmp0 (g_getenv ("XDG_CURRENT_DESKTOP"), "Unity") == 0) | 11:23 |
| alkisg | The || part is Ubuntu-specific | 11:23 |
| alkisg | In upstream it's plain (n_sources < 2) | 11:24 |
| alkisg | That's what I'm asking to revert to what it is in upstream | 11:24 |
| alkisg | It's the same in unity-settings-daemon too | 11:24 |
| seb128 | that doesn't make sense | 11:24 |
| seb128 | oh | 11:25 |
| alkisg | It says, "remove the grp: if we're running Unity" | 11:25 |
| seb128 | attente, is that supposed to be a && ? | 11:25 |
| attente | seb128, should be an || | 11:25 |
| attente | seb128, alkisg, not sure what the consequences of removing it are | 11:26 |
| attente | alkisg, have you tested without it? | 11:26 |
| seb128 | why did we add it? | 11:26 |
| alkisg | attente: I only tried with gnome-fallback and XDG_CURRENT_DESKTOP=Gnome unity-settings-daemon --replace &, then I came here to ask about why it was added, to see what to test or find other workarounds for them... :) | 11:27 |
| attente | seb128, i added it because i didn't want xkb to switch layouts when unity+u-s-d were already handling it | 11:27 |
| alkisg | Let me say some related things: | 11:27 |
| attente | but if it's inconsequential and only fixes things, then i'm all for removing it | 11:27 |
| darkxst | alkisg, XDG_CURRENT_DESKTOP=Gnome is not correct, you are just getting nothing behaviour with that | 11:27 |
| mdeslaur | seb128: hi! yes, I'll apply the fix on top of the SRU | 11:27 |
| alkisg | darkxst: I just wanted to skip the "|| Unity" check | 11:27 |
| seb128 | mdeslaur, hey, thanks! | 11:27 |
| alkisg | To switch from us,gr, I'm using either (1) Win+Space, the gnome/unity defined shortcut, or (2) Alt+Shift, the XKB-defined shortcut. | 11:28 |
| seb128 | alkisg, that leads to the issue you described though "if you cycle using xkb, the indicator is wrong"? | 11:28 |
| darkxst | alkisg, its XDG_CURRENT_DESKTOP=GNOME | 11:28 |
| alkisg | seb128: right, but I don't know how, it's solved in Fedora 20 | 11:28 |
| mdeslaur | seb128: thanks! Sorry for the mail, but I was going Eod and wanted you to see it before I got back | 11:28 |
| seb128 | mdeslaur, no worry, emails are good, please keep sending some in such cases ;-) | 11:28 |
| alkisg | I'll try to locate that after we agree on the first one, i.e. that keyboard switching in fullscreen applications is too important to drop | 11:29 |
| seb128 | alkisg, so how do users configure/known about alt-shift? | 11:29 |
| seb128 | isn't that confusing? | 11:29 |
| seb128 | shouldn't win-space work in all cases? | 11:29 |
| alkisg | It's the default in xorg, when choosing "greek" in ubiquity | 11:29 |
| seb128 | (though I guess it can't due to the xorg grabs limitation) | 11:29 |
| alkisg | And it affects the console too | 11:29 |
| seb128 | you happen to know about it | 11:29 |
| alkisg | So, the gnome solution isn't really good, it breaks a lot of stuff | 11:30 |
| seb128 | but it's not displayed in the UI | 11:30 |
| alkisg | $ grep OPTIONS /etc/default/keyboard | 11:30 |
| alkisg | XKBOPTIONS="grp:alt_shift_toggle,grp_led:scroll" | 11:30 |
| xnox | seb128: i've just commented on bug #1242572 ubiquity sets up alt+shift for layout switching, since console-setup does not support super+space combo yet. | 11:30 |
| ubot2 | Launchpad bug 1242572 in ubiquity (Ubuntu Trusty) "Ubiquity sets Alt+Shift shortcut for layout switching, while installed system uses Super+Space" [Wishlist,Confirmed] https://launchpad.net/bugs/1242572 | 11:30 |
| seb128 | xnox, ok | 11:30 |
| xnox | i'm going to send a bug report to debian to start supporting super+space which is modern universal layout switcher. | 11:31 |
| alkisg | So when xorg starts, it starts with Alt+Shift. Then, Gnome also adds Win+Space. Then, unity-settings-daemon blocks Alt+Shift, and that last part is what I'm trying to get fixed first. | 11:31 |
| xnox | alkisg: is XKBOPTIONS something Xorg parsers? | 11:31 |
| seb128 | alkisg, so you get 2 different keybindings doing the same thing through different mechanism, isn't that confusing? | 11:31 |
| alkisg | Yes | 11:31 |
| xnox | alkisg: and if yes, then what's the combo for super_space? or a list of all supported combos? | 11:31 |
| alkisg | seb128: people don't yet know Super+Space, it's a new thing | 11:32 |
| xnox | alkisg: super+space is default in unity, gnome3, windows7&8, mac os x. | 11:32 |
| alkisg | From the windows world, up to windows vista it was Alt+Shift, same as in xorg, | 11:32 |
| seb128 | alkisg, people do know about super-space, that's what other OSes use (e.g win or macOS) and that's what is displayed in the settings UI | 11:32 |
| alkisg | And I think in windows 7/8 (I don't use them), the started Win+Space | 11:32 |
| seb128 | right | 11:32 |
| alkisg | So, to sum up: | 11:33 |
| alkisg | Gnome broke the ability to get notified from xkb when I switch layouts in fullscreen apps, and doesn't want patches about that. | 11:34 |
| alkisg | I.e. in fedora 20, in tuxtype, I can press alt+shift to switch languages, but when I exit, the indicator doesn't get updated | 11:34 |
| alkisg | *didn't | 11:34 |
| alkisg | That the _only_ part that doesn't work OK in fedora 20, everything else is fine. It's pretty minor. | 11:34 |
| alkisg | I can press alt+shift OR win+space and change layout while in gnome-shell. They both make the indicator update itself | 11:35 |
| seb128 | alkisg, alt-shift is a leftover not exposed in an user visible way, you just happen to know about it it seems | 11:35 |
| alkisg | In ubuntu, even if I manage to get the alt_shift_toggle back, the indicator doesn't get updated. That's my second thing I want to help solve. | 11:35 |
| alkisg | seb128: me and everyone that has been using it for the last 20 years... :) | 11:36 |
| alkisg | We don't know about Win+Space... that's the new thing | 11:36 |
| alkisg | And, it doesn't work in tuxtype, we're not able to type Greek there in 14.04, while we were able to in 12.04 | 11:36 |
| seb128 | alkisg, sorry to tell you that, but you are a minority | 11:36 |
| alkisg | seb128: I'm talking about *all* Greeks | 11:37 |
| alkisg | And possibly many other countries | 11:37 |
| seb128 | alkisg, there are more users of win7/8/macOS/unity/gnome-shell than users of 20 years old unix | 11:37 |
| seb128 | alkisg, I'm speaking about not knowing super-space | 11:37 |
| alkisg | seb128: http://www.ltsp.org/stories/widget-map/?location=Greece | 11:37 |
| xnox | seb128: reading man xkeyboard-config there is no support for grp:lwin_space_toggle =( | 11:37 |
| alkisg | None of the schools there has anything newer than windows xp | 11:37 |
| alkisg | 1000+ schools, multiply with a few labs each... | 11:37 |
| attente | so are the full screen applications doing a full active grab of the keyboard? | 11:37 |
| seb128 | attente, seems they do | 11:38 |
| alkisg | Yes | 11:38 |
| alkisg | seb128: I don't mind about alt+shift or win+space. What I mind about more is, being able to type greek in tuxpaint/tuxtype/etc etc | 11:39 |
| seb128 | alkisg, anyway that discussion about people not knowing super-space is not useful, please just stop assuming that's true/important | 11:39 |
| seb128 | alkisg, right | 11:39 |
| seb128 | alkisg, so please stop trying to make a case against the keybinding ;-) | 11:39 |
| alkisg | The window manager can't help there, it can't see the switch event | 11:39 |
| alkisg | I wan't, sorry if it sounded like I started that part... | 11:39 |
| alkisg | *wasn't | 11:39 |
| seb128 | no worry | 11:39 |
| attente | i do wonder what happens if we remove that condition and just try what upstream does | 11:40 |
| seb128 | do you know if there is a bug in launchpad for the tux issue? | 11:40 |
| attente | sounds racy | 11:40 |
| alkisg | attente: I think we'll get bug reports about "alt+shift doesn't update the indicator", which is the second more significant thing that I'd like to work on | 11:40 |
| seb128 | in which way? | 11:40 |
| seb128 | well, ideally we wouldn't have 2 ways | 11:40 |
| alkisg | attente: It does work in Fedora though, so I'm confident there's some solution there provided upstream... | 11:41 |
| seb128 | but it seems like due to xorg it's either the group cycle workaround to let those users to have a way to cycle | 11:41 |
| seb128 | or nothing | 11:41 |
| alkisg | seb128: fedora does have 2 ways too | 11:41 |
| seb128 | right | 11:41 |
| alkisg | And, console will have alt+shift for some time... | 11:41 |
| seb128 | due to xorg I guess | 11:41 |
| alkisg | Right | 11:41 |
| seb128 | ideally we would have 1 channel | 11:41 |
| seb128 | working for everything | 11:41 |
| seb128 | I guess the gnome-shell indicator listen for the xorg grp changes and react to them to update its config or something | 11:42 |
| seb128 | attente, is indicator-keyboard doing that? | 11:42 |
| alkisg | No, it isn't, I tried it | 11:42 |
| alkisg | It's listening for a gsettings change | 11:42 |
| alkisg | /org/gnome/desktop/input-sources/current | 11:43 |
| attente | yeah, alkisg is right | 11:43 |
| seb128 | which means that if we remove that u-s-d condition, we would create issues for users for hit alt-shift | 11:43 |
| seb128 | like some could do it without noticing | 11:43 |
| alkisg | Create some, solve some | 11:43 |
| seb128 | and get an invalid keymap | 11:43 |
| seb128 | or at least not matching the indicator | 11:43 |
| alkisg | The ones solved are more important. And I'm confident the others can be solved as well. | 11:43 |
| seb128 | not sure they are more important | 11:44 |
| xnox | mlankhorst: can xkeyboard-config support something like "grp:lwin_space_toggle" ? as in "switch keyboard layouts using Super + Space" ? This is for bug #1242572 | 11:44 |
| ubot2 | Launchpad bug 1242572 in xkeyboard-config (Ubuntu Trusty) "xkeyboard-config, console-setup, and ubiquity should use Super+Space for switching keyboard layouts" [Wishlist,Confirmed] https://launchpad.net/bugs/1242572 | 11:44 |
| seb128 | alkisg, what happen if you switch to greek before starting the game? you can then input in greek? | 11:44 |
| mlankhorst | xnox: no idea, what happens when you try? | 11:44 |
| alkisg | seb128: yes. Well, in some other cases, like teeworlds, I can't play the game, because it expects a latin "a" for left, instead of a greek "α" | 11:45 |
| alkisg | seb128: this also works: sleep 20 && dbus-send --session --type=method_call --print-reply --reply-timeout=2000 --dest=org.gnome.SettingsDaemon.Keyboard /org/gnome/SettingsDaemon/Keyboard org.gnome.SettingsDaemon.Keyboard.SetInputSource uint32:1 & tuxpaint --fullscreen | 11:45 |
| alkisg | I.e. it changes the layout while inside the fullscreen app | 11:46 |
| seb128 | right | 11:46 |
| seb128 | well the issue is that the game grabs the keyboard | 11:46 |
| alkisg | Right | 11:46 |
| seb128 | so you super-space doesn't reach compiz anymore | 11:46 |
| seb128 | I don't see a way around using the xorg cycling while we use X | 11:46 |
| alkisg | That's a xorg limitation; the problem is that gnome, window managers etc, haven't thought that limitation when designing the new way to manage layouts | 11:46 |
| seb128 | it's not they didn't | 11:47 |
| seb128 | it's that they designed a clean architecture, that is going to work fine for wayland | 11:47 |
| alkisg | But anyway that's an upstream decision, no point in discussing it here unless we want to carry a patch that translates xkb events => gsettings | 11:47 |
| alkisg | They're using backends | 11:47 |
| alkisg | xkb is one of the backends | 11:47 |
| alkisg | xkb provides a hook | 11:47 |
| alkisg | They just chose not to allow other implementations change the layout, and use that hook to get notified | 11:48 |
| alkisg | They decided that only gnome would use xkb... | 11:48 |
| alkisg | Anyway, not what we should be talking about | 11:48 |
| attente | alkisg, it would also depend on what keyboard layout they start the application with too | 11:48 |
| alkisg | So, what do we do now? | 11:48 |
| attente | alkisg, if the user starts with a US layout, then they still won't be able to switch to the Greek layout because it won't exist in the user's layouts | 11:49 |
| xnox | mlankhorst: good point. i wonder how to test it without settings-daemon/compiz/etal getting in the way. | 11:49 |
| alkisg | Is the "can't type greek inside fullscreen apps" enough of a problem, to drop the "|| desktop=unity" code above? | 11:49 |
| seb128 | alkisg, it's enough of a problem to deserve being resolved, we are unsure that dropping the code is the best way to resolve it (yet) | 11:50 |
| alkisg | OK, let's see what issues that would cause | 11:50 |
| alkisg | 1) 2 ways to change layouts | 11:50 |
| alkisg | That the same in fedora, and I think also in windows | 11:50 |
| attente | seb128, alkisg, we might be able to do it, but we also need to add a bit of code to unity to handle the ISO_Next_Group key code | 11:50 |
| alkisg | Let me check in windows... | 11:50 |
| attente | alkisg, but that still won't solve the problem if the user starts the app in US | 11:51 |
| alkisg | attente: it will | 11:51 |
| alkisg | Windows 7 doesn't support win+space, only alt+shift | 11:51 |
| attente | alkisg, switch to a US layout, then try 'setxkbmap -query' | 11:51 |
| alkisg | attente: with xorg, we never switch layouts, it's always "us,gr" | 11:52 |
| alkisg | We switch... I don't know the terminology... the active set? | 11:52 |
| alkisg | The xprop never gets updated | 11:52 |
| attente | alkisg, is unity-settings-daemon not running for you? | 11:52 |
| attente | unity-settings-daemon overrides the layouts lists when the input source is changed | 11:52 |
| alkisg | If I change that with win+space, then it gets to "gr,us" | 11:52 |
| * alkisg tries to say that better: with xorg, it's always: "us,gr". with gnome, it's: "us,gr" or "gr,us" | 11:53 | |
| alkisg | It's never plain "us" | 11:53 |
| attente | alkisg, what i'm seeing now is not the same | 11:53 |
| === alan_g is now known as alan_g|lunch | ||
| attente | if i'm using unity-control-center, adding us and gr, then switch to us, i only see us when i call setxkbmap | 11:54 |
| mlankhorst | xnox: murder them :D | 11:54 |
| mlankhorst | or run a diff flavor | 11:54 |
| attente | alkisg, because every time u-s-d switches input sources, it uploads a new xkb configuration | 11:55 |
| alkisg | attente: it does, and it's "us,gr" or "gr,us"... now I'm with gnome-fallback, want me to switch to unity? | 11:55 |
| xnox | mlankhorst: oh does like xubuntu/lubuntu actually use X to switch layouts? I've tried $ startx gedit, and i can't switch keyboard layout in it. | 11:55 |
| attente | alkisg, i'm using unity and that's what's happening for me | 11:56 |
| mlankhorst | xnox: not sure | 11:56 |
| attente | alkisg, please try it :) | 11:56 |
| alkisg | attente: ah. There's also a note in the input-sources schema, let me find it... | 11:56 |
| attente | although i don't really understand why it wouldn't be the same, isn't unity-settings-daemon also running under gnome-fallback? | 11:57 |
| alkisg | It is | 11:57 |
| alkisg | attente: what's the output of this? gsettings get org.gnome.desktop.input-sources sources | 11:57 |
| alkisg | [('xkb', 'us'), ('xkb', 'gr')] for me | 11:57 |
| attente | anyways, we can try to remove it, but we do need to add something to unity | 11:57 |
| attente | alkisg, [('xkb', 'us'), ('xkb', 'gr')] | 11:58 |
| attente | and switching to 'us' results in only 'us' for the xkb configuration | 11:58 |
| alkisg | attente: do you have ibus running? | 11:59 |
| alkisg | If so, could you stop it? | 11:59 |
| alkisg | That's another bug, it shouldn't be running at all, but that's for upstream im-config... | 11:59 |
| alkisg | Tooo many keyboard-related bugs in the 12.04 => 14.04 transision.. :( | 11:59 |
| attente | alkisg, killing ibus doesn't change what i'm seeing | 12:03 |
| alkisg | OK give me a few minutes to test with unity | 12:04 |
| attente | ok | 12:05 |
| alkisg | But, I think that the implementation parts are the details, if we could agree first on what we want to do, I'm pretty sure we could provide patches for that. | 12:06 |
| alkisg | What I'd like to do, is make Ubuntu 14.04 do what Fedora 20 does, which isn't perfect like Ubuntu <= 12.04 was, but it's good enough. Everything works, even the indicator gets updated when I press alt+shift, and the only minor detail that doesn't work is that the indicator doesn't get updated while I'm inside a full screen app that grabs the keyboard | 12:06 |
| alkisg | If we could agree that we want to accept patches in that direction, I'm pretty sure it's implementable without diverting from upstream | 12:06 |
| * alkisg checks unity... | 12:06 | |
| seb128 | the goal sounds fine yes | 12:06 |
| attente | so remove the condition from u-s-d, handle ISO_Next_Group in unity, then maybe upload a secondary layout in u-s-d? | 12:07 |
| attente | sounds like it should be ok | 12:08 |
| alkisg | attente: if you have patches for that, I'd be more than glad to extensively test them :) | 12:08 |
| attente | alkisg, sure :) | 12:09 |
| attente | actually... i guess it's all u-s-d | 12:09 |
| xnox | mlankhorst: so i've definately tested xorg now, stopped lightdm / quit all unity sessions, did "startx", launched terminal, kill all gnome/unity-settings-deamons, then use setxkbmap to get russian and usa layout. | 12:10 |
| xnox | mlankhorst: lwin_space_toggle, win_space_toggle do not work, win_toggle does work. | 12:10 |
| attente | alkisg, have you checked setxkbmap under unity? | 12:10 |
| xnox | mlankhorst: so yeah, x clearly does not support super_space here. | 12:10 |
| mlankhorst | seems so :P | 12:10 |
| xnox | mlankhorst: can you forward that bug upstream? or should I file it somewhere? | 12:10 |
| alkisg | attente: I'm creating a livecd, because I've changed my system too much to be sure that it's clean... | 12:10 |
| mlankhorst | but does it work with unity? | 12:10 |
| xnox | mlankhorst: with unity - either gnome-settings-daemon or unity-settings-daemon or ibus or compiz intercept "super+space" and switch layouts. | 12:11 |
| xnox | mlankhorst: however this does not work e.g. in full-screen x apps, etc. as widely discussed above. | 12:11 |
| attente | alkisg, it's ok, no matter. i already know that it doesn't work on my machine, so that's something else to fix | 12:11 |
| mlankhorst | xnox: ok | 12:13 |
| xnox | mlankhorst: registering in the freedesktop bugzilla to open a bug report. | 12:14 |
| xnox | mlankhorst: filed https://bugs.freedesktop.org/show_bug.cgi?id=78076 | 12:20 |
| ubot2 | Freedesktop bug 78076 in General "Please add support for grp:lwin_space_toggle and similar" [Enhancement,New] | 12:20 |
| xnox | mlankhorst: would you write a patch for that? :-) | 12:23 |
| Laney | attente: check these beasts crushing hard https://www.youtube.com/watch?v=xAB9-VGIkzM | 12:24 |
| mdeslaur | Trevinho, seb128: http://www.ubuntu.com/usn/usn-2184-1/ | 12:26 |
| mdeslaur | Trevinho, seb128: thanks! :) | 12:26 |
| seb128 | mdeslaur, thanks! | 12:27 |
| alkisg | Thank you very much guys, I'll be back later on to try to help as much as I can. :) | 12:27 |
| mlankhorst | xnox: suddenly everyone wants something from me :P | 12:27 |
| === alkisg is now known as work_alkisg | ||
| seb128 | work_alkisg, thanks for pointing those issues and helping getting them resolved! | 12:27 |
| attente | Laney, impressed to see any free soloing ;) | 12:29 |
| Laney | definitely bouldering opportunities in malta btw ;-) | 12:30 |
| Laney | (including DWS...) | 12:30 |
| attente | sounds fun :) | 12:30 |
| Laney | scary | 12:31 |
| xnox | mlankhorst: how would like to be bribed? =) | 12:37 |
| mlankhorst | ok lets see.. | 12:40 |
| mlankhorst | xnox: looks like it should be fixable easily :P | 12:43 |
| mlankhorst | copy alt_space_toggle I guess | 12:44 |
| mlankhorst | hm no slightly more involved | 12:45 |
| xnox | Laney: with updated gnome-keyring job, it appears as if launchpadlib based things (e.g. like any ubuntu-archive / ubuntu-dev-tools scripts) fail to work, time out activating secrets api and fail to store lp token in the keyring. | 12:54 |
| xnox | Laney: have you noticed this? | 12:54 |
| Laney | no I use those inside lxc | 12:54 |
| Laney | lemme try | 12:54 |
| Laney | xnox: yeah ... | 12:58 |
| Laney | that's supposed to be dbus activated | 12:58 |
| === MacSlow|lunch is now known as MacSlow | ||
| === alan_g|lunch is now known as alan_g | ||
| Laney | xnox: I think it's because it's not on the user's bus | 13:06 |
| xnox | Laney: so gnome-keyring must start on started dbus? | 13:06 |
| Laney | try it, that made it work for me | 13:07 |
| Laney | then we get back to gpg/ssh ordering problems again | 13:07 |
| xnox | yeah it's racing with dbus job which is starting on xsession-init. | 13:08 |
| xnox | do we? i thought we wouldn't. gpg-/ssh- agents will not start, until gnome-keyring completes even if it's blocked by dbus not started yet. | 13:08 |
| * xnox will add sleeps to simulate blockage. | 13:09 | |
| Laney | oh, you mean that started dbus and <what we have now> will work? | 13:09 |
| xnox | yeah. | 13:10 |
| xnox | "started dbus and (<what we have>)" | 13:10 |
| xnox | that's what i'm testing here now. | 13:10 |
| Laney | k | 13:10 |
| Laney | lunch! | 13:14 |
| Laney | i'm thinking fried eggs | 13:14 |
| mlankhorst | xnox: is it ok if both super keys work? :P | 13:14 |
| xnox | mlankhorst: yes, that's even better cause that's how it currently works under gnome-settings-daemon | 13:14 |
| seb128 | mlankhorst, is http://paste.ubuntu.com/7359309/ having what you need? | 13:16 |
| mlankhorst | looks good, but grr @ xf86platformVTProbe crap | 13:19 |
| mlankhorst | xnox: I have no idea what I'm doing, so here it is http://paste.debian.net/96403/ :P | 13:23 |
| mlankhorst | will probably break or something | 13:23 |
| xnox | mlankhorst: let me test it =) | 13:23 |
| mlankhorst | or fail to load | 13:24 |
| xnox | mlankhorst: better that, than the experimental efibootmgr i'm about to test. | 13:24 |
| xnox | mlankhorst: edit comments "toggle using lwin + space as combo" to say it's "win + space", not "lwin + space". | 13:25 |
| seb128 | mlankhorst, do you want me to copy that somewhere, or are you going to send to upstream/to whatever bug you use for the issue? | 13:26 |
| mlankhorst | seb128: nah just checking at this point, going to need hardware first to do a bisect | 13:27 |
| seb128 | k | 13:27 |
| seb128 | the valgrind error/bt are not enough for upstream to have a clue? | 13:27 |
| mlankhorst | no they can reproduce that already | 13:28 |
| seb128 | so they should be able to fix it? | 13:28 |
| mlankhorst | oh that vt probe bug I already fixed in utopic | 13:28 |
| mlankhorst | but the deliveremulatedmotionevent looks weird | 13:29 |
| seb128 | the log has another error from the touch | 13:29 |
| mlankhorst | yeah looking at that, seems weird | 13:32 |
| mlankhorst | it shouldn't be called beyond end of the event queue, so the bug is not in EnqueueEvent, I think. | 13:34 |
| xnox | mlankhorst: loads correctly, does not work. space acts like a space, instead of "win+space" combo. | 13:36 |
| xnox | (when pressing simultaniously or holding down windows key and then pressing spacebar) | 13:36 |
| xnox | let me just try one more time, just in case. | 13:37 |
| mlankhorst | hm was afraid of that :P | 13:37 |
| mlankhorst | xnox: what if you load meta_win too? | 13:40 |
| xnox | altwin:meta_win ? | 13:40 |
| xnox | mlankhorst: loading: setxkbmap -model pc105 -layout us,ru -option grp:win_space_toggle,altwin:meta_win | 13:41 |
| xnox | makes windows+space not produce a " ", but it doesn't change layout either. | 13:41 |
| mlankhorst | hm | 13:41 |
| mlankhorst | I guess the mod4 should be Meta | 13:41 |
| xnox | oh, loading altwin:meta_win rmeoves win_space_toggle. | 13:42 |
| mlankhorst | ok, can you try the patch with Mod4 changed to Meta ? | 13:43 |
| xnox | mlankhorst: ok. | 13:44 |
| mlankhorst | I think the altwin one should no longer be needed then | 13:50 |
| xnox | mlankhorst: doesn't work, with or without. | 13:50 |
| mlankhorst | gr, stupid xkb-data defining a whole language | 13:51 |
| mlankhorst | xnox: http://paste.debian.net/96410/ | 13:56 |
| xnox | mlankhorst: i wonder if all my tests are flawed, since i'm running them under compiz and e.g. Super+S does compiz wall thing. | 14:03 |
| === oCrazyLem is now known as CrazyLemon | ||
| xnox | *sigh* | 14:04 |
| xnox | still no. | 14:04 |
| mlankhorst | :/ | 14:05 |
| mlankhorst | well I have no clue anyway, just guessing based on the syntax from xkb-data | 14:05 |
| xnox | i'll try again later in something more basic, like lubuntu. | 14:06 |
| mlankhorst | try a raw xserver + xev | 14:12 |
| mdeslaur | seb128: I need a bit of help... | 14:20 |
| mdeslaur | seb128: http://bazaar.launchpad.net/~indicator-applet-developers/indicator-datetime/trunk.13.10/revision/282 | 14:21 |
| mdeslaur | seb128: that causes unity-greeter to segfault, and I can<t seem to figure out why | 14:21 |
| larsu | mdeslaur: do you have a backtrace? | 14:26 |
| mdeslaur | larsu: not yet, I haven't figured out how to get one from the greeter | 14:29 |
| === alan_g is now known as alan_g|tea | ||
| mdeslaur | seb128: hrm, this work though (but is a hack...) http://paste.ubuntu.com/7359767/ | 14:42 |
| larsu | mdeslaur: that also removes functionality.... | 14:44 |
| mdeslaur | larsu: oh? | 14:45 |
| larsu | mdeslaur: you unsetting an action that is referenced by the menu item. charles will know what it does exactly | 14:46 |
| mdeslaur | larsu: I added the else, the old code simply didn:t set activation-action | 14:47 |
| === alan_g|tea is now known as alan_g | ||
| mdeslaur | charles: could you take a look? | 14:47 |
| larsu | mdeslaur: ah right, I misread. Probably a bug in the menu item itself then (which is in lp:ido) | 14:48 |
| charles | mdeslaur, sure | 14:48 |
| mdeslaur | charles: for context, I'm trying to release http://bazaar.launchpad.net/~indicator-applet-developers/indicator-datetime/trunk.13.10/revision/282 as a security update | 14:48 |
| mdeslaur | charles: but that commit causes the greeter to segfault | 14:49 |
| mdeslaur | seems something doesn't like not having "activation-action" set...but I can't figure out why | 14:49 |
| charles | mdeslaur, do you have a stacktrace? | 14:50 |
| mdeslaur | charles: how can I get a stacktrace for unity-greeter? (that's what segfaults) | 14:51 |
| charles | good question, I don't know. larsu? | 14:51 |
| larsu | charles: no clue, but you should be able to reproduce it with the loader (it includes the greeter profile) | 14:52 |
| mdeslaur | here;s my log, fwiw: http://paste.ubuntu.com/7359836/ | 14:53 |
| mdeslaur | larsu: the loader? | 14:53 |
| charles | that's a good thought | 14:54 |
| charles | mdeslaur, install libindicator3-tools and try running "indicator-loader3 com.canonical.indicator.datetime" | 14:54 |
| charles | and see if that crashes or ont | 14:54 |
| mdeslaur | ok, one sec | 14:55 |
| charles | indicator-loader3 will show you all the profiles' menus for a service | 14:55 |
| charles | so if that crashes too, it's a quick way to get a stacktrace | 14:55 |
| mdeslaur | oh, that<s pretty cool | 14:57 |
| larsu | mdeslaur: rather /usr/lib/x86_64-linux-gnu/indicator-loader3 /usr/share/unity/indicators/com.canonical.indicator.datetime | 14:58 |
| larsu | I can't reproduce it here, though | 14:58 |
| larsu | ah, maybe because I don't have e-d-s set up | 14:58 |
| mdeslaur | apt-get install evolution | 14:58 |
| larsu | ya ... still can't reproduce | 14:58 |
| mdeslaur | hrm, how do I make it load my updated package... | 14:59 |
| Trevinho | seb128: hi | 14:59 |
| larsu | seb128: he's out doing exersise | 15:00 |
| larsu | *exercise | 15:00 |
| Trevinho | larsu: ah, ok thanks | 15:00 |
| mdeslaur | ah! got it | 15:01 |
| larsu | can you reproduce the crash? | 15:04 |
| mlankhorst | xnox: ? | 15:05 |
| larsu | charles: from the rough stacktrace in mdeslaur's paste, it looks like one of the qdata is set incorrectly | 15:05 |
| larsu | qdata uses g_data_list | 15:05 |
| xnox | mlankhorst: moved on to debugging efibootmgr instead =) | 15:07 |
| mdeslaur | larsu: I can, yes, by double-clicking a date in the calendar...I'm trying to get a proper backtrace now | 15:09 |
| charles | larsu, that's plausible from the log, but at first read I don't see how we get to a qdata error from that diff -- it's not using qdata directly, nor indirectly with strings | 15:11 |
| mdeslaur | this isn't helpful at all: http://paste.ubuntu.com/7359979/ | 15:11 |
| Laney | aww, was diffing e-d-s against the wrong left version | 15:11 |
| larsu | charles: yes it is, it calls set_data_full for activation-action and selection-action | 15:11 |
| Laney | "holy shit, this can be a sync?" | 15:11 |
| mdeslaur | larsu: what calls set_data_full for activation-action? | 15:16 |
| charles | mdeslaur, so you're getting this by applying http://bazaar.launchpad.net/~indicator-applet-developers/indicator-datetime/trunk.13.10/revision/282 to Saucy, yes? | 15:16 |
| mlankhorst | seb128: I might not make the meeting, in that case: testing vt switching and filing/fixing bugs related to it. investigating some touch bugs, fixing bug 1301839, some upstream kernel work | 15:16 |
| ubot2 | Launchpad bug 1301839 in xserver-xorg-video-intel (Ubuntu) "AMD Radeon HD 8670A/8670M/8690M[1002:6660] Low-graphic mode with Trusty" [Undecided,Confirmed] https://launchpad.net/bugs/1301839 | 15:16 |
| mlankhorst | and bug 1313986 | 15:16 |
| ubot2 | Launchpad bug 1313986 in linux (Ubuntu) "[GeForce 9400 GT] Nouveau driver does not work with kernel 3.13.0-24 (Ubuntu 14.04)" [Undecided,Incomplete] https://launchpad.net/bugs/1313986 | 15:16 |
| mdeslaur | charles: yes, and running in the greeter | 15:16 |
| larsu | mdeslaur: http://bazaar.launchpad.net/~indicator-applet-developers/ido/trunk.14.04/view/head:/src/idocalendarmenuitem.c#L623 | 15:16 |
| charles | mdeslaur, and using that formula you're also getting the crash in indicator-loader3? | 15:16 |
| mdeslaur | charles: yes, I had to log into my session, update my datetime packages, kill the datetime daemon in my session, and then use indicator-loader3 | 15:17 |
| mdeslaur | charles: indicator-loader3 crashes when I select the greeter one, and then double click on a date or two | 15:18 |
| charles | mdeslaur, ok I'll try to reproduce here | 15:18 |
| charles | larsu, good point about set_data_full there | 15:18 |
| charles | so you're right, we are using qdata, I wasn't thinking of ido | 15:19 |
| charles | but the crash shows it is coming from inside libgtk, so that makes sense | 15:19 |
| larsu | you set the qdata on a widget and the crash happens when it is destroyed | 15:20 |
| mdeslaur | charles: fyi, I did this during testing earlier just to see: http://paste.ubuntu.com/7359767/ | 15:20 |
| brookswarner | KombuchaKip - you here for Desktop meeting? | 15:21 |
| seb128 | Trevinho, hey | 15:26 |
| Trevinho | seb128: Hi! | 15:27 |
| Trevinho | I was wondering ... since https://code.launchpad.net/~3v1n0/unity/lockscreen-keys-disable/+merge/217528 is in archive trough the mdeslaur's distro patch, can we now manually sync trunk merging that? | 15:27 |
| seb128 | Trevinho, why manually? didn't you merge back when you landed it? | 15:29 |
| seb128 | Trevinho, is it still in a silo ? | 15:29 |
| seb128 | in which case you can m&c the silo | 15:29 |
| seb128 | oh, it's meeting time | 15:30 |
| Trevinho | seb128: well, bregma is not here this week thus we (as a team) don't have any way to get a silo | 15:30 |
| Trevinho | seb128: mh, ok, talk later | 15:30 |
| seb128 | qengho, Sweetshark_gc, mlankhorst, Laney, tkamppeter, desrt, attente, larsu, kenvandine: hey, it's meeting time! | 15:31 |
| * Sweetshark_gc readjusts his beach towel. | 15:31 | |
| tkamppeter | seb128, can I start this time, as I have to leave earlier. | 15:31 |
| charles | mdeslaur, which branch should I try applying that patch to, in order to re-create the crash? | 15:31 |
| desrt | OMeetinG | 15:31 |
| seb128 | tkamppeter, sure, go ahead | 15:31 |
| tkamppeter | - Installed Utopic on one machine for development | 15:32 |
| tkamppeter | - cups-filters: Released 1.0.53 with several security fixes | 15:32 |
| tkamppeter | - ghostscript: Fix to make it work when incompatible fonts are installed | 15:32 |
| tkamppeter | - system-config-printer: Suppress running HPLIP checks on non-HP printers | 15:32 |
| tkamppeter | - HPLIP: Fix for HP OfficeJet Pro K550 which is EOL at HP. | 15:32 |
| tkamppeter | - Bugs. | 15:32 |
| * qengho grumbles. | 15:32 | |
| mdeslaur | charles: the package in saucy...I don<t think there;s a branch for it | 15:32 |
| desrt | mdeslaur: (meeting) | 15:32 |
| seb128 | tkamppeter, did you find anyone to ping about the libspecte/ghostscript issue? | 15:33 |
| kenvandine | hey seb128, sorry i'm in a call atm | 15:34 |
| seb128 | kenvandine, no worry, some time before your turn, or you can skip if needed | 15:34 |
| tkamppeter | seb128, I have looked into it, it is definitely not Ghostscript, but for sure libspectre. So we have to wait for Marek Kasik to comment on the Freedesktop bug. I have commented on the LP bug. | 15:35 |
| seb128 | ok | 15:35 |
| KombuchaKip | brookswarner: I'm here buddy. I'm always idling in here. | 15:35 |
| seb128 | tkamppeter, do you have contacts with marek? maybe you could try pinging him? | 15:35 |
| seb128 | tkamppeter, thanks for looking to the issue in any case | 15:36 |
| tkamppeter | seb128, yes, I will send an e-mail directly to him then. | 15:36 |
| seb128 | tkamppeter, thanks | 15:36 |
| seb128 | ok, next is qengho | 15:36 |
| seb128 | qengho, hey | 15:36 |
| qengho | Hey! | 15:36 |
| qengho | - in-progress: Fixing the problems aura brings to chromium, one at a time: horiz | 15:37 |
| qengho | ontal scroll, widget placement when DPI adjusted, menu sensitivity | 15:37 |
| qengho | - to-do: tab screwwyness bug. | 15:37 |
| qengho | - in-progress: Trying to get 34.0.1847.131 out, which should fix a few bugs. (This week's way upstream makes my life interesting: Deprecating make as build tool.) | 15:37 |
| qengho | EOF | 15:37 |
| KombuchaKip | seb128: If you or anybody else would like to test my upstream patch for Mozilla Thunderbird, that would be awesome. https://bugzilla.mozilla.org/show_bug.cgi?id=824909 | 15:37 |
| ubot2 | Mozilla bug 824909 in OS Integration "can't print .eml files - print preview remains blank" [Normal,Assigned] | 15:37 |
| seb128 | KombuchaKip, hum, please wait for your turn ;-) | 15:37 |
| seb128 | qengho, do you know if the update fixes the "loose tabs from the session/mix order" issue? | 15:38 |
| qengho | seb128: I don't know if it does. | 15:38 |
| seb128 | ok | 15:38 |
| charles | mdeslaur, what command line would I use in Trusty to get the version of i-datetime that you're patching against? | 15:38 |
| seb128 | I guess it's not the only bug on your list ;-) | 15:38 |
| seb128 | charles, (we are in a meeting, please use the other channel where we moved the discussion) | 15:39 |
| charles | ack | 15:39 |
| seb128 | thanks | 15:39 |
| seb128 | qengho, thanks | 15:39 |
| seb128 | mlankhorst, there? | 15:39 |
| KombuchaKip | seb128: Yeah man, I can imagine. Bugs galore. | 15:40 |
| seb128 | ok, he said he might not be there anymore | 15:40 |
| seb128 | his status update was | 15:40 |
| seb128 | " I might not make the meeting, in that case: testing vt switching and filing/fixing bugs related to it. investigating some touch bugs, fixing bug 1301839, some upstream kernel work | 15:40 |
| ubot2 | Launchpad bug 1301839 in xserver-xorg-video-intel (Ubuntu) "AMD Radeon HD 8670A/8670M/8690M[1002:6660] Low-graphic mode with Trusty" [Undecided,Confirmed] https://launchpad.net/bugs/1301839 | 15:40 |
| seb128 | and bug 1313986 | 15:40 |
| seb128 | " | 15:40 |
| ubot2 | Launchpad bug 1313986 in linux (Ubuntu) "[GeForce 9400 GT] Nouveau driver does not work with kernel 3.13.0-24 (Ubuntu 14.04)" [Undecided,Incomplete] https://launchpad.net/bugs/1313986 | 15:40 |
| seb128 | ok | 15:41 |
| seb128 | Sweetshark_gc, you are next ;-) | 15:41 |
| Sweetshark_gc | - LibreOffice Hackfest Las Palmas 2014 https://wiki.documentfoundation.org/Hackfest/GranCanaria2014 | 15:41 |
| Sweetshark_gc | -- lightning talks in front of students | 15:41 |
| Sweetshark_gc | -- wrote a unittest for writer | 15:41 |
| Sweetshark_gc | -- investigated mail merge wizard issues, unfortunately quite a mess | 15:41 |
| Sweetshark_gc | -- various meetings | 15:41 |
| Sweetshark_gc | - TDF administrative stuff | 15:41 |
| Sweetshark_gc | - various mails and coordination | 15:41 |
| Sweetshark_gc | - now back to that LibreOffice unity menu issue :/ | 15:41 |
| Sweetshark_gc | EOF | 15:41 |
| seb128 | Sweetshark_gc, when do you fly back? | 15:41 |
| Sweetshark_gc | next wednesday. Ill take friday of and thursday is labour day, so three more work days. | 15:42 |
| Sweetshark_gc | I hope to find an opportunity to catch up with the aentos guys (who wrote the unity integration stuff) for dinner or something still ... | 15:43 |
| seb128 | ok | 15:43 |
| seb128 | have fun there! | 15:43 |
| seb128 | (same in France btw, thursday is off) | 15:43 |
| Laney | :o | 15:44 |
| seb128 | good luck for the menus issue as well, let me know when you get a fix, seems like a SRU topic | 15:44 |
| larsu | (same in Germany, but I'll be at the hackfest anyway) | 15:44 |
| seb128 | Sweetshark_gc, thanks | 15:44 |
| seb128 | Laney, your turn | 15:44 |
| Laney | • distro-info patch / discussions to not fail so hard when the info gets out of date (due to not having a codename on release day which happens every time now) | 15:44 |
| Laney | • webkit SRU / update | 15:44 |
| Laney | • gstreamer SRU / update | 15:44 |
| Laney | • apport SRU for str vs bytes bug in precise | 15:44 |
| Laney | • Update vala to 0.24 & make it the default | 15:44 |
| Laney | • Various merges and syncs from merges.u.c, including (prepared, not uploaded) cheese 3.12 with de-headerbarification patch | 15:44 |
| Laney | • Discussions / testing an upstart job for gnome-keyring to resolve the races (others please test the package in U) | 15:44 |
| Laney | • Start looking at e-d-s 3.12 | 15:44 |
| Laney | • Brief systemd test, basically boots a system but my /e/n/i interfaces didn't come up :( | 15:44 |
| Laney | ⚠ | 15:44 |
| Laney | oh I'm off Monday and Tuesday (to continue the holidays theme) | 15:45 |
| seb128 | Laney, I saw your cheese patches got commited upstream, good job ;-) | 15:45 |
| Laney | ya, amigadave is a good guy | 15:46 |
| seb128 | Laney, the 1st is not a bank holiday in the u.k? | 15:46 |
| Laney | no | 15:46 |
| Laney | monday is though | 15:46 |
| ogra_ | weirdos | 15:46 |
| ogra_ | :P | 15:46 |
| Laney | 'early may bank holiday' | 15:46 |
| seb128 | k | 15:46 |
| seb128 | enjoy that one ;-) | 15:47 |
| ogra_ | go with the rest of the world :) | 15:47 |
| seb128 | Laney, thanks | 15:47 |
| seb128 | desrt, your turn | 15:47 |
| === ChrisTownsend1 is now known as ChrisTownsend | ||
| desrt | hihi | 15:47 |
| desrt | in berlin this week for the gtk hackfest | 15:47 |
| Laney | they moved it from may 1st, don't know why (some places like schools have festivities on that day anyway) | 15:48 |
| desrt | did some work on adding support for horizontal rows of buttons in gmenumodel -- made sure we have a good fallback for if we try to display it in a normal menu | 15:48 |
| desrt | (ie: we make it as a section with a hint -- if we don't support the hint, we will just display the items normally) | 15:48 |
| desrt | did some work on better reporting of invalid dbus messages (due to bugs in gmenumodel client that were crashing hud) | 15:48 |
| desrt | also added support for Implements= in gdesktopappinfo | 15:48 |
| desrt | the usual bug fixing, etc. | 15:49 |
| desrt | oh... spent some time looking into the issues about headerbar/csd and resizing under compiz | 15:49 |
| desrt | this stuff is pretty complicated, but there is at least a partial gtk bug here... but it has nothing to do with the resizing | 15:49 |
| desrt | it seems that compiz doesn't support resize on windows with borders but no headerbar.... and that's a really really old hint that we ought to support | 15:50 |
| desrt | Trevinho: i think you said you might be able to help here? | 15:50 |
| desrt | maybe not. that's all. | 15:51 |
| seb128 | let's keep the discussion with Trevinho as an after meeting topic | 15:51 |
| Trevinho | desrt: yes, it's all in unity's DecoratedWindow code | 15:51 |
| seb128 | I would like to discuss a bit GTK/GNOME updates at the end of the meeting anyway | 15:52 |
| seb128 | desrt, thanks | 15:52 |
| seb128 | attente, hey | 15:52 |
| larsu | update all the things \o/ | 15:52 |
| desrt | larsu: s/all/ALL/ | 15:52 |
| larsu | not that many... | 15:52 |
| Laney | I already almost typoed a couple of times | 15:52 |
| attente | seb128, hi | 15:52 |
| Laney | luckily I typoed ppa:laney/u-merges instead of ubuntu | 15:53 |
| desrt | Laney is fishing for beer in malta... | 15:53 |
| Laney | baa | 15:53 |
| * larsu would definitely chime in on beer for Laney | 15:53 | |
| attente | i spent more time debugging eclipse menus, uploaded to a ppa | 15:53 |
| larsu | *chip in | 15:53 |
| attente | also looked into non-latin shortcuts for java/inkscape/blender | 15:54 |
| attente | i uploaded that to a ppa, but it breaks keyboard layout switching under gnome-shell, and causes regressions under unity if the layout switcher is set to shift | 15:55 |
| seb128 | do you need more testers for some of those? | 15:56 |
| larsu | some russian testers? | 15:56 |
| attente | i think the eclipse fix is ok, but really need to figure out what's happening for the non-latin shortcuts | 15:57 |
| desrt | attente: gtk does some internal mapping to figure out the equivalent latin code | 15:57 |
| desrt | depending on the set of all installed keymaps | 15:58 |
| desrt | i had to track this down back in the day when i wrote that altgrabber stuff... | 15:58 |
| attente | desrt, yeah | 15:58 |
| attente | but this is for java/inkscape/blender | 15:58 |
| attente | which are all assuming a latin keyboard group 0 | 15:58 |
| desrt | so even if you're in english mode your hebrew shortcuts will work for menus written in hebrew, based on which physical key you press | 15:58 |
| seb128 | same issue as libreoffice? | 15:58 |
| attente | seb128, yes | 15:58 |
| seb128 | shrug | 15:58 |
| desrt | ah. tricky. | 15:58 |
| seb128 | some days I hate those GNOME changes | 15:59 |
| attente | so i just swapped the order | 15:59 |
| attente | in u-s-d and g-s-d | 15:59 |
| seb128 | things were much easier/robust when we were using xkb | 15:59 |
| attente | in g-s-d this works, but breaks the xkb grp: option | 15:59 |
| desrt | seb128: you only hate gnome changes on _some_ days? :) | 15:59 |
| larsu | only on days ending in '-day' | 15:59 |
| seb128 | ;-) | 15:59 |
| desrt | larsu: and wednesday | 15:59 |
| attente | in u-s-d this works, unless you set your switching shortcut to left shift right shift, or both | 15:59 |
| seb128 | shift+something you mean? | 16:00 |
| attente | seb128, well. to be more accurate, it causes a regression where you cannot type capital letters in the non-latin layout | 16:00 |
| seb128 | that seems annoying indeed | 16:01 |
| attente | ppa is here: https://launchpad.net/~attente/+archive/java-non-latin-shortcuts | 16:01 |
| seb128 | in the libreoffice bug Rui (the GNOME guy who did the changes) basically said that libreoffice needs to be "fixed" to not assume that group 0 is a latin keyboard | 16:02 |
| seb128 | so might be that we need to "fix" java as well? | 16:02 |
| attente | i think that's unrealistic to patch every package doing this | 16:02 |
| seb128 | attente, what's the approach you are taking to try to make it work? | 16:02 |
| seb128 | right, I think so too | 16:03 |
| attente | seb128, basically reversing the order as set by g/u-s-d | 16:03 |
| seb128 | which is why I'm unhappy about those GNOME changes :/ | 16:03 |
| seb128 | to have a latin back in group 0? | 16:03 |
| attente | so instead of ru,us, you get us,ru | 16:03 |
| attente | seb128, yes | 16:03 |
| seb128 | why does that create issues for shift? | 16:03 |
| attente | if you set your keyboard layout switcher to shift | 16:04 |
| desrt | maybe details for later? | 16:05 |
| attente | unity grabs that so that it can do a modifier-only grab on shift-<random character> | 16:05 |
| larsu | +1 | 16:05 |
| seb128 | yes, I was starting thinkg that | 16:05 |
| attente | yeah... | 16:05 |
| attente | sorry | 16:05 |
| seb128 | no worry, I'm asking for some details because we have everybody around | 16:05 |
| seb128 | which includes desrt | 16:05 |
| larsu | right | 16:05 |
| seb128 | but I guess we can discuss that again later | 16:05 |
| seb128 | attente, thanks | 16:05 |
| desrt | LARSU | 16:05 |
| desrt | GO! | 16:06 |
| seb128 | larsu, your turn | 16:06 |
| * larsu doesn't listen to anyone but seb128 | 16:06 | |
| larsu | oh. shit. | 16:06 |
| seb128 | lol | 16:06 |
| larsu | it was a bug fixing week again | 16:06 |
| larsu | some minor evince issues (some accels stopped working after the gmenu port) | 16:06 |
| larsu | some playing around with the sound widgets in ido | 16:07 |
| larsu | which is getting SRUed | 16:07 |
| seb128 | did you stack those evince fixes/did them in the same branch btw? | 16:07 |
| desrt | larsu: we're getting a slider in upstream gmenumodel soon. i'd like your input there. | 16:07 |
| seb128 | I've those on my list of things to look at/land | 16:07 |
| larsu | seb128: no, they're in different branches but should be fairly independent | 16:07 |
| larsu | let me know if they don't merge cleanly | 16:08 |
| seb128 | ok, good | 16:08 |
| larsu | desrt: yes, let's talk about that later though | 16:08 |
| desrt | ofc | 16:08 |
| larsu | also, gmenumodel was crahsing for the hud when the hud sent invalid dbus messages to it | 16:08 |
| larsu | I added some input validation checks | 16:08 |
| larsu | (desrt was talking about this too, he added g_dbus_warning() for this reason) | 16:09 |
| desrt | larsu: if you review g_dbus_warning() i'll review your stuff that uses it :p | 16:09 |
| larsu | will do | 16:09 |
| larsu | had some conversations with desrt about the headerbar issue | 16:09 |
| larsu | I think that's about it... crazy week with lots of small things | 16:10 |
| seb128 | larsu, thanks | 16:11 |
| seb128 | kenvandine, there? | 16:11 |
| seb128 | likely not | 16:12 |
| seb128 | so my turn | 16:12 |
| seb128 | * spent most of the week in launchpad, reading the bugs filed after release/triaging/targetting the ones to SRU | 16:12 |
| seb128 | * did some SRUs (empathy, rhythmbox) | 16:12 |
| seb128 | * helped testing some other SRUs | 16:12 |
| seb128 | * some debugging (valgrined xorg for touch/xserver segfault issues) | 16:12 |
| seb128 | * quite some IRC discussions (gtkheaderbar, new gnome-desktop, GNOME updates) | 16:12 |
| seb128 | </week> | 16:12 |
| vedic | Which gui do you recommend? I have tried xfce4 but I think I need a bit better but light as well. xfce4 is light but I need a bit better | 16:12 |
| seb128 | KombuchaKip, oh, sorry I almost forgot you ... you had something to share I think | 16:13 |
| desrt | vedic: (desktop team meeting in progress right now, sorry) | 16:13 |
| seb128 | vedic, hey, try #ubuntu for user questions | 16:13 |
| KombuchaKip | seb128: Hey man. No problem. Yes, a patch. https://bugzilla.mozilla.org/show_bug.cgi?id=824909 | 16:13 |
| ubot2 | Mozilla bug 824909 in OS Integration "can't print .eml files - print preview remains blank" [Normal,Assigned] | 16:13 |
| seb128 | did you manage to get it in a ppa for testing? | 16:13 |
| vedic | seb128,, desrt: Ok. Thanks | 16:14 |
| seb128 | KombuchaKip, did you manage to get it in a ppa for testing? | 16:14 |
| KombuchaKip | seb128: No not going to do that yet until upstream accepts it. Then I'll build a PPA, and then I'll SRU it. | 16:15 |
| seb128 | ok | 16:15 |
| KombuchaKip | seb128: Otherwise there is no point in going through all of that if upstream says it's no good. | 16:15 |
| seb128 | well, if upstream accept it/merge it you probably don't need to SRU | 16:15 |
| seb128 | we security update new versions | 16:15 |
| seb128 | right | 16:15 |
| KombuchaKip | seb128: Yep | 16:15 |
| seb128 | good luck with that, seems like they have been busy with releases but have it on their review list | 16:16 |
| seb128 | KombuchaKip, thanks | 16:16 |
| desrt | and back to seb | 16:16 |
| seb128 | ok, so end of roundtable | 16:16 |
| larsu | GNOME! | 16:16 |
| seb128 | let's discuss GNOME updates | 16:16 |
| larsu | glib: yes. gtk: yes. | 16:16 |
| desrt | i know larsu's opinion | 16:16 |
| KombuchaKip | seb128: No problem. So far they seem agreeable upstream and I can't see why they'd turn it down. Looks like there isn't much being done on Thunderbird these days, so you'd think they'd take it. | 16:16 |
| larsu | I know desrt's. | 16:16 |
| seb128 | desrt, what is planned in glib this cycle ? | 16:16 |
| larsu | GWallclock | 16:16 |
| desrt | seb128: nothing major | 16:17 |
| larsu | EggObjectList *cough* | 16:17 |
| desrt | probably will get the mainloop done this cycle... | 16:17 |
| seb128 | desrt, same as usual anyway, should be stable and if it's not you are there to fix the issues, right? | 16:17 |
| desrt | in terms of new APIs.... it seems quiet right now | 16:17 |
| larsu | desktop file cache? | 16:17 |
| desrt | things have moved back to gtk hacking more... | 16:17 |
| desrt | larsu: ya.. maybe this as well, but again, not new API | 16:17 |
| larsu | right | 16:17 |
| seb128 | ok | 16:17 |
| desrt | seb128: ya.... glib is pretty low-risk, for sure | 16:17 |
| seb128 | so glib 2.41/42: ack | 16:17 |
| desrt | can't speak to gtk yet... i suspect we will know more after this week | 16:18 |
| seb128 | gtk 3.12: ack | 16:18 |
| larsu | I think we should get the new gtk. I suspect that we can get rid of a lot of the background hacks I've had to put into the theme | 16:18 |
| seb128 | (seems to be easy, still need some testing before landing though) | 16:18 |
| larsu | because 3.12 has the unified drawing stuff | 16:18 |
| seb128 | larsu, new = 3.12 right? | 16:18 |
| desrt | pixelcache++ | 16:18 |
| larsu | I'll find this out in the next days | 16:18 |
| seb128 | not 3.13 | 16:18 |
| larsu | seb128: yes | 16:18 |
| seb128 | +1 for that | 16:18 |
| larsu | we'll bother you about 3.13 once 3.12 is in+ | 16:18 |
| seb128 | I plan to throw it to the desktop team ppa for some extra testing first | 16:18 |
| seb128 | right | 16:18 |
| larsu | great | 16:18 |
| seb128 | let's settle down on the Debian merges, etc first | 16:19 |
| larsu | I've been running it for a while today and it seems to cause no major issues | 16:19 |
| desrt | would be nice to follow unstable with the base libraries as well | 16:19 |
| Laney | we'll need to keep an eye on dialog widths | 16:19 |
| seb128 | then we can discuss 3.13 maybe | 16:19 |
| desrt | but maybe wait for the second half of the cycle | 16:19 |
| Laney | and I think there's something with dialog headers too | 16:19 |
| desrt | soup, dconf, gvfs, etc, etc | 16:19 |
| seb128 | right | 16:19 |
| larsu | Laney: yes, dialog headers are client-side now | 16:19 |
| larsu | which is stupid, but I wouldn't consider it a major issue | 16:19 |
| Laney | isn't there a gtksetting for it? | 16:19 |
| larsu | there should be ya | 16:19 |
| desrt | there may be some things we find upsetting about the new gtkdialog behaviours | 16:19 |
| desrt | like how the buttons are in message dialogs now... kinda ugly imho | 16:20 |
| desrt | they fill the entire bottom of the dialog | 16:20 |
| larsu | isn't that fixable with css? | 16:20 |
| desrt | i doubt it... | 16:20 |
| desrt | there is only limited control over sizing from css... | 16:20 |
| desrt | stuff like padding/margins/etc | 16:20 |
| larsu | margin? | 16:20 |
| larsu | now there's no margin - setting one should put the buttons back to were they where, no? | 16:20 |
| seb128 | is that in 3.12 ? do we have an example/screenshot of what we are talking about? | 16:21 |
| desrt | no... | 16:21 |
| larsu | anyhow, we can have a look | 16:21 |
| desrt | https://wiki.gnome.org/Design/HIG/Dialogs has mockups | 16:21 |
| desrt | pretty much looks like this | 16:21 |
| seb128 | k | 16:22 |
| seb128 | well anyway let's get the new gtk in the ppa and see how things are going with it | 16:22 |
| desrt | http://blogs.gnome.org/mclasen/files/2014/03/gedit-messagedialog.png is the real deal | 16:22 |
| desrt | not so nice.... | 16:22 |
| seb128 | urg, indeed not | 16:22 |
| desrt | probably not too difficult to fix, but it would be some patching | 16:22 |
| desrt | not just theme | 16:22 |
| larsu | who's idea was that anyway? | 16:22 |
| desrt | designers | 16:23 |
| seb128 | yeah, and "why"? | 16:23 |
| desrt | seb128: nfc. | 16:23 |
| seb128 | but yeah, I think we need to fix that | 16:23 |
| larsu | well, it's 3.13 material | 16:23 |
| seb128 | with some luck they get feedback that make them reconsider it | 16:23 |
| larsu | let's fix it when we get therew | 16:23 |
| seb128 | +1 | 16:23 |
| seb128 | larsu, but are you sure? | 16:23 |
| seb128 | larsu, that screenshot has "for 3.12" written in the gedit in the screenshot | 16:24 |
| larsu | oh, might be ya | 16:24 |
| larsu | we need to fix it, then :D | 16:24 |
| seb128 | agreed | 16:24 |
| seb128 | ok | 16:24 |
| seb128 | so glib->2.42 | 16:24 |
| seb128 | gtk->3.12 | 16:24 |
| desrt | also note: the other things get turned off via xsettings | 16:24 |
| desrt | so the headerbar dialogs will not be on on unity, by default | 16:24 |
| seb128 | likely going to update gvfs/libsoup/librsvg/gdk-pixbuf/g-i as usual | 16:25 |
| larsu | they are right now though | 16:25 |
| seb128 | headerbar dialog default to off iirc | 16:25 |
| larsu | seb128: the buttons are in the bottom, but the titlebar is still csd | 16:25 |
| desrt | ya... we had quite an argument about this | 16:25 |
| larsu | something messed up there | 16:25 |
| larsu | *is | 16:25 |
| seb128 | k, something to watch for | 16:26 |
| seb128 | so summary | 16:26 |
| seb128 | * glib->2.42, gtk->3.12 | 16:26 |
| seb128 | * likely going to update gvfs/libsoup/librsvg/gdk-pixbuf/g-i as usual | 16:26 |
| seb128 | * desrt/larsu/Trevinho working on make csd work better under Unity | 16:26 |
| desrt | i'll chat with benjamin tomorrow about that | 16:26 |
| seb128 | we are waiting on the outcome of ^ to determine what to do with apps | 16:27 |
| desrt | LIM is problematic... | 16:27 |
| Laney | what about the apps we de-csded? | 16:27 |
| seb128 | default would be to update to 3.12/merge with Debian for things not using GtkHeaderBar | 16:27 |
| desrt | seb128: not too many more of those around... | 16:27 |
| seb128 | let's stay away from adding GtkHeaderBar uses until we know better where we stand | 16:27 |
| seb128 | Laney, like? | 16:27 |
| seb128 | Laney, well, as said ^, I would suggest staying away from CSD until we have a better idea on how well it works | 16:28 |
| seb128 | desrt has a good point with LIM | 16:28 |
| Laney | yes, for apps where it's not a problem | 16:28 |
| Laney | like evolution | 16:28 |
| seb128 | Laney, it is a problem for evolution? | 16:28 |
| larsu | we could have LIM with csd - we wouldn't even need dbus! | 16:28 |
| desrt | putting the 'local' in LIM | 16:29 |
| * desrt likes it | 16:29 | |
| larsu | LLIM | 16:29 |
| larsu | locally locally integrated menus | 16:29 |
| * desrt llikes it | 16:29 | |
| seb128 | larsu, not consistent with LIM in other apps though | 16:29 |
| larsu | seb128: it woudld be visually | 16:29 |
| seb128 | like display on mouseover? | 16:29 |
| larsu | *it could be | 16:29 |
| larsu | seb128: well, we'd need to add code for that obviously | 16:29 |
| desrt | seb128: dunno if you notice but anything is possible in gtk these days :p | 16:29 |
| seb128 | lol | 16:30 |
| larsu | except setting multiple accels for an action | 16:30 |
| seb128 | well, one thing at the time | 16:30 |
| Laney | is it? | 16:30 |
| seb128 | let's get CSD working in Unity first | 16:30 |
| desrt | larsu: i said 'these days', not 'last year' | 16:30 |
| larsu | desrt: 'last year' is what we have in ubuntu though | 16:30 |
| desrt | s/CSD/resizing of windows with no titlebars/ | 16:30 |
| seb128 | well, we also need to have a non duplicated decoration | 16:31 |
| desrt | ya... this is the motif hint | 16:31 |
| desrt | you can request a border (ie: resizable) but not a titlebar | 16:31 |
| desrt | this is the one that falls all over the place | 16:31 |
| desrt | in unity this means you get neither | 16:31 |
| larsu | compiz needs to get with the times! Think of all the motif apps not working | 16:31 |
| desrt | in kde/xfce you get both | 16:31 |
| desrt | anyway.... won't change the world | 16:32 |
| seb128 | k | 16:33 |
| larsu | this meeting is getting longer than it used to be | 16:33 |
| seb128 | well, we know what to work on/have enough decided to start the cycle | 16:33 |
| larsu | right | 16:33 |
| seb128 | larsu, well, over 1h is wrong for sure | 16:33 |
| seb128 | but we had the extra topic | 16:34 |
| seb128 | so let's wrap there | 16:34 |
| seb128 | we can discuss more specifics/updates later in the cycle | 16:34 |
| larsu | ya, not complaining, just observing | 16:34 |
| seb128 | there is a vUDS in june as well | 16:34 |
| desrt | (it's not true. he was complaining.) | 16:34 |
| seb128 | we should have enough to keep busy until there | 16:34 |
| larsu | (desrt is lying) | 16:34 |
| desrt | i think we're done :) | 16:34 |
| desrt | i'll buy larsu an icecream to make him happy | 16:34 |
| === alan_g is now known as alan_g|EOD | ||
| larsu | \o/ | 16:34 |
| seb128 | larsu wants his kebab | 16:34 |
| larsu | not now | 16:35 |
| seb128 | ;-) | 16:35 |
| seb128 | Laney, did you have anything more you wanted to discuss there? | 16:35 |
| attente | so... about the java/inkscape/blender shortcuts... | 16:35 |
| Laney | not really, if non-headerbar apps are ok to update to | 16:35 |
| seb128 | Laney, or is the "let's update glib/gtk, deal with Debian merges, make csd work better" then figure out what we do next a good start of cycle plan for you? | 16:35 |
| seb128 | yes | 16:35 |
| seb128 | let's stay away from adding for headerbars until we work more on that | 16:36 |
| seb128 | but everything else is fine to go 3.12 | 16:36 |
| Laney | ok | 16:36 |
| seb128 | thanks everyone | 16:36 |
| seb128 | desrt, larsu: please help attente! | 16:36 |
| Laney | we should maybe work on empathy/telepathy/IM at some point ;-) | 16:36 |
| seb128 | attente, I'm unsure if you were trolling them btw :p | 16:36 |
| attente | lol | 16:36 |
| Laney | (or use pidgin) | 16:36 |
| seb128 | Laney, use pidgin! | 16:37 |
| desrt | larsu has already closed his laptop, with prejudice | 16:37 |
| attente | seb128, about the eclipse fix, i think it's ok for an sru | 16:38 |
| attente | well, unity-gtk-module fix for eclipse | 16:38 |
| seb128 | attente, did you mp it already ? | 16:38 |
| attente | seb128, https://code.launchpad.net/~attente/unity-gtk-module/1208019-2/+merge/216964 | 16:38 |
| seb128 | attente, thanks | 16:40 |
| seb128 | desrt, larsu, charles, tedg: ^ can you review that one? | 16:40 |
| seb128 | in looks fine to me in principle | 16:40 |
| Laney | seb128: are you going to upload that 3.12 to the desktop ppa? | 16:40 |
| seb128 | Laney, yes, but probably not this week, rather next (since thursday is an holiday and I'm taking friday off) | 16:41 |
| desrt | seb128: larsu and i are just about to head out for respective dinner dates | 16:41 |
| Laney | ok i'll upload it to u-merges then | 16:41 |
| seb128 | Laney, if you want to do it feel free | 16:41 |
| Laney | not going to review the diff right now | 16:41 |
| Laney | I just want it atm to build some stuff | 16:41 |
| seb128 | Laney, u-merges? | 16:42 |
| Laney | my 'u' testing ppa | 16:42 |
| seb128 | desrt, larsu: have fun! | 16:42 |
| seb128 | Laney, oh, ok | 16:42 |
| ogra_ | oh man ... this pile of lockscreen bugs is a PR disaster | 16:50 |
| Laney | erm | 16:51 |
| ogra_ | (is there any key combo that doesnt let you bypass the lock ? ) | 16:51 |
| Laney | trolling? | 16:51 |
| ogra_ | every time i open a new news site there was a new one found in LP ... and they pick on all of them | 16:51 |
| ogra_ | Laney, nope | 16:52 |
| ogra_ | alt-f2 ... not ctrl-alt-t | 16:52 |
| ogra_ | s/not/now/ | 16:52 |
| seb128 | ogra_, it's 1 bug | 16:52 |
| ogra_ | is it ? | 16:53 |
| seb128 | yes | 16:53 |
| ogra_ | bug 1313885 suggests ctrl-alt-t is new | 16:53 |
| ubot2 | Launchpad bug 1313885 in unity (Ubuntu Utopic) "lock screen bypass" [Critical,In progress] https://launchpad.net/bugs/1313885 | 16:53 |
| seb128 | ogra_, it's the same issue | 16:53 |
| ogra_ | but a different fix ? | 16:54 |
| seb128 | ogra_, it's "by playing with indicator you can end up having the keyboard focus in the session" | 16:54 |
| seb128 | no | 16:54 |
| ogra_ | ah, k | 16:54 |
| seb128 | the fix is buggy/incomplete | 16:54 |
| seb128 | you can still end up with the keyboard going to the session | 16:54 |
| seb128 | which let you do "keybinding of your choice" | 16:54 |
| ogra_ | well, in any case news pages seem to love to pick it up | 16:55 |
| seb128 | well, nothing we can't do about that, out of fixing the issues | 16:56 |
| seb128 | can't->can | 16:56 |
| seb128 | ogra_, stop reading so much "news" sites | 16:57 |
| mdeslaur | ogra_: feel free to send this list of bugs to the news sites: https://bugs.launchpad.net/ubuntu/+source/gnome-screensaver?field.searchtext=&orderby=-datecreated | 16:57 |
| === Ursinha is now known as Ursinha-afk | ||
| ogra_ | seb128, lol | 16:59 |
| Laney | It's what you get if you do everything in the open | 17:00 |
| ogra_ | mdeslaur, after fighting with two reporters for half the night about bug 1308572 i gave up doing such stuff | 17:00 |
| ubot2 | Launchpad bug 1308572 in unity (Ubuntu) "Ubuntu 14.04: security problem in the lock screen" [Critical,Fix released] https://launchpad.net/bugs/1308572 | 17:00 |
| ogra_ | they nicely reported it as present in the release in their news | 17:01 |
| mdeslaur | ogra_: apparently reporting about bugs that are already fixed doesn't bring readers :) | 17:01 |
| ogra_ | mdeslaur, "Ubuntu 14.04 released with serious lockscreen bug !!!111oneone" makes a wonderful clickbait headline | 17:02 |
| Laney | I wouldn't get worked up about it | 17:03 |
| Laney | Bugs happen, they get fixed, we move on | 17:03 |
| ogra_ | yeah | 17:03 |
| mdeslaur | ogra_: I publish 2-5 USNs every single week, they do realize there are a constant flow of security issues, right? :) | 17:03 |
| Laney | If someone wants to make a few cents writing about them then that's up them as far as I'm concerned | 17:04 |
| Laney | it's not worth getting distracted over | 17:04 |
| mdeslaur | anyway, /me goes back to work | 17:04 |
| ogra_ | mdeslaur, i would expect they do ... after all its the biggest german security news site ... (though i would also expect some serious research from them, its the first time i see them doing such a clickbait thing) | 17:05 |
| === Ursinha-afk is now known as Ursinha | ||
| asac | with bregma being on vacation who would know something about how the unity8 session with mir works? | 18:28 |
| asac | on deslktop? | 18:28 |
| asac | and lightdm etc. | 18:28 |
| seb128 | asac, mterry might be able to help you | 19:01 |
| * mterry looks up | 19:01 | |
| mterry | I don't see scrollback | 19:01 |
| seb128 | <asac> with bregma being on vacation who would know something about how the unity8 session with mir works? | 19:02 |
| seb128 | on deslktop? | 19:02 |
| seb128 | and lightdm etc. | 19:02 |
| seb128 | mterry, ^ | 19:02 |
| mterry | asac, I might yeah | 19:02 |
| seb128 | mterry, sorry, just random shoot, you seem like one of those who could know | 19:02 |
| tedg | larsu, I thought we had a signal for when a menu was shown if there was a special action defined, no? | 19:07 |
| asac | mterry: Saviq is already talking to stgraber in -touch on this topic now... lets see if they figure it :) | 19:10 |
| Saviq | mterry, come to -touch, you'll definitely be helpful | 19:10 |
| larsu | tedg: we do, see the "submenu-action" section in https://wiki.gnome.org/Projects/GLib/GApplication/GMenuModel | 22:56 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!
