| [00:26] <RAOF> robert_ancell: Heh. Launchpad timeout. If that's the bug I'm thinking of, no, I didn't forget to upstream it. | |
| [00:27] <RAOF> robert_ancell: colord upstream has totally removed sane support due to crashes; if it looks like my fix clears most of those crashes up sane support will be re-added. | |
| [02:09] <thumper> hi all | |
| [02:09] <thumper> seems my GM45 intel chip still isn't very loved by quantal | |
| [02:10] <bkerensa> that is most unfortunate | |
| [02:10] <bkerensa> what are you encountering? | |
| [02:11] <thumper> after logging in from lightdm, unity starts, then after 10s it crashes and restarts, then crashes again after 10s and doesn't restart | |
| [02:11] <thumper> 10s delay is the time when the dash view gets realized | |
| [02:11] <thumper> I filed a bug a while back | |
| [02:11] <thumper> and it is a known issue with the older intel chips | |
| [02:11] <bkerensa> thumper: what is the bug #? | |
| [02:11] <bkerensa> oh | |
| [02:11] <thumper> some race condition in the driver | |
| [02:11] * thumper looks | |
| [02:13] <thumper> https://bugs.launchpad.net/ubuntu/+source/xserver-xorg-video-intel/+bug/1043562 | |
| [02:13] <ubot2> Launchpad bug 1043562 in xserver-xorg-video-intel "[gm45] GPU lockup EIR: 0x00000010 PGTBL_ER: 0x00000001 IPEHR: 0x60020100" [High,Triaged] | |
| [02:20] <Darxus> I should be updating my gtk ppa to this version, right https://launchpad.net/ubuntu/quantal/+source/gtk+3.0/3.5.18-0ubuntu2 ? Not 3.5.18-0ubuntu1 which I can currently download from the archives? | |
| [03:54] <pitti> Good morning | |
| [03:59] <TheMuso> Morning pitti. | |
| [04:00] <pitti> hey TheMuso, how are you? | |
| [04:00] <TheMuso> pitti: Not too bad thanks, yourself? | |
| [04:01] <pitti> quite fine, thanks! | |
| [04:01] <pitti> hey robert_ancell | |
| [04:01] <robert_ancell> pitti, hello | |
| [04:02] <pitti> robert_ancell: how's glib doing? | |
| [04:02] <robert_ancell> pitti, i'm building it in a PPA right now, doesn't like to build locally | |
| [04:03] * TheMuso wonders whether GTK 3.5.16 needs the new glib... I am experiencing some segfaults related to a11y and GTK here, downgrading to 3.5.16 makes the problem go away... | |
| [04:03] <pitti> robert_ancell: you can build locally with DEB_BUILD_OPTIONS=nocheck if you want local .debs for testing (and quick) | |
| [04:03] <TheMuso> Just building a downgaded package of the first ubuntu revision of 3.5.18 prior to the patch update... | |
| [04:03] <TheMuso> to see if the problem still persists. | |
| [04:03] <robert_ancell> pitti, oh, that's handy | |
| [04:04] <pitti> robert_ancell: this dbus keyring thingy is really unnerving; I don't know why it does that when building under a running gnome session | |
| [04:04] <pitti> robert_ancell: I looked for something which would create that key (then we coudl add it to debian/rules), or to avoid it (unsetting DBUS_*, etc.), but no luck | |
| [04:04] <jbicha> robert_ancell: you saw that glib 2.33.14 was uploaded to experimental today? | |
| [04:05] <pitti> oh, we might just sync that then? | |
| [04:05] <pitti> seems Laney committed it there | |
| [04:06] <pitti> https://buildd.debian.org/status/package.php?p=glib2.0&suite=experimental | |
| [04:06] <pitti> wow | |
| [04:06] <robert_ancell> just sync then | |
| [04:10] <pitti> oh, this also disables the glib schema compiler warnings again, good | |
| [04:10] <pitti> d'oh, forgot to specify -proposed | |
| [04:10] <pitti> ah well, I'll watch it | |
| === attente_ is now known as attente | |
| [05:23] <didrocks> good morning | |
| [05:26] <pitti> bonjour didrocks | |
| [05:27] <didrocks> hey pitti, how are you? | |
| [05:29] <pitti> je vais bien, merci! et toi? | |
| [05:29] <kenvandine> didrocks, good morning | |
| [05:29] <kenvandine> didrocks, is the libunity release going to be 6.5.2? | |
| [05:29] <pitti> kenvandine: oh, burning the midnight oil? | |
| [05:29] <kenvandine> yes... sigh | |
| [05:29] <kenvandine> i am taking today off... so gotta get some stuff finished :) | |
| [05:30] <kenvandine> so i can go out of town in a few hours | |
| [05:30] <didrocks> kenvandine: 6.6 rather | |
| [05:30] <kenvandine> ok | |
| [05:30] <kenvandine> cool | |
| [05:30] <didrocks> pitti: je continue de tousser encore :/ | |
| [05:30] <kenvandine> didrocks, i released gwibber | |
| [05:30] <kenvandine> and preparing the packaging branch | |
| [05:30] <kenvandine> going to add a build dep for that | |
| [05:31] <pitti> didrocks: argh -- no time to rest and cure it properly, I guess :( | |
| [05:31] <didrocks> kenvandine: excellent! pushing to -proposed? | |
| [05:31] <kenvandine> no... | |
| [05:31] <kenvandine> oh | |
| [05:31] <kenvandine> i guess i can | |
| [05:31] <kenvandine> it'll just dep wait :) | |
| [05:31] <didrocks> kenvandine: I think it will make sense :) | |
| [05:31] <kenvandine> i was going to ask you to upload it later | |
| [05:31] <kenvandine> but this works :) | |
| [05:31] <didrocks> can do as well :) | |
| [05:31] <didrocks> kenvandine: indicators -> all is ok? | |
| [05:31] <kenvandine> can you get chromium-browser out of -proposed for me? | |
| [05:31] <kenvandine> yeah | |
| [05:32] <didrocks> pitti: yeah, the rush will continue today | |
| [05:32] <didrocks> kenvandine: will do as well! :) | |
| [05:32] <kenvandine> didrocks, dbusmenu is still coming | |
| [05:32] <kenvandine> thx | |
| [05:32] <didrocks> kenvandine: ok, excellent, thanks! | |
| [05:32] <kenvandine> it is ready to go from -proposed to quantal now | |
| [05:32] * didrocks continues to do changes in unity now that the latest UIFe/FFe are acked | |
| [05:32] <kenvandine> bamf and libunity-webapps got promoted | |
| [05:33] <kenvandine> i am going to make gwibber build dep on libunity-dev >= 6.5.1 (which is what i have in the pkgconfig check) | |
| [05:38] <RAOF> kenvandine: Is https://bugs.launchpad.net/ubuntu/+source/gwibber/+bug/1051801 on your plate? | |
| [05:38] <ubot2> Launchpad bug 1051801 in gwibber "gwibber-service crashed with SIGSEGV" [Medium,Confirmed] | |
| [05:39] <didrocks> kenvandine: looking good to me | |
| [05:39] <kenvandine> RAOF, yes... that is a dupe though | |
| [05:39] <kenvandine> i am currently stumped on that crash | |
| [05:39] <kenvandine> i had barry look at it too, no joy :/ | |
| [05:40] <RAOF> kenvandine: You're probably corrupting memory in libdbus by not calling dbus_threads_init_default | |
| [05:41] <kenvandine> RAOF, do you think that could be in libaccounts-glib and triggered by gwibber calling it with GI? | |
| [05:42] <didrocks> RAOF is the libdbus expert now I guess :) | |
| [05:42] <kenvandine> we need one :) | |
| [05:43] <didrocks> RAOF: more colord stuff on the release team list for you man :) | |
| [05:45] <RAOF> Hurray! :( | |
| [05:45] <kenvandine> didrocks, don't distract him from thinking about that crash :) | |
| [05:45] <didrocks> heh :) | |
| [05:45] <didrocks> ok rls-q-incoming clean for this morning now | |
| [05:48] <didrocks> kenvandine: hum, you did upload unity-lens-photos? | |
| [05:48] <didrocks> kenvandine: it was part of the testing stack for unity, as it delivers new features | |
| [05:48] <didrocks> did you get all manual tests done? | |
| [05:49] <kenvandine> there was no new features | |
| [05:49] <didrocks> oh -photos | |
| [05:49] <didrocks> not -videos | |
| [05:49] <kenvandine> :) | |
| [05:49] * didrocks will grab coffee | |
| [05:49] <didrocks> kenvandine: go back to do some awesomeness, sorry for the interruption ;) | |
| [05:49] <kenvandine> no worries | |
| [05:50] <RAOF> kenvandine: In the stacktrace on that bug, there's something that's accessing dbus from python; that's going to break, because gdbus accesses dbus on its own thread. | |
| [05:51] <kenvandine> gwibber-service isn't doing any gdbus yet | |
| [05:52] <RAOF> glib *always* does gdbus. | |
| [05:52] <kenvandine> but... it is using libaccounts-glib | |
| [05:52] <kenvandine> well.. sure :) | |
| [05:52] <RAOF> It's frustrating. It means that glib has broken previously-working code. | |
| [05:53] <kenvandine> i am reasonable sure it is something to do with logging in with libaccounts-glib | |
| [05:53] <kenvandine> we stubbed that out so it didn't actually call libaccounts-glib and couldn't get the crash | |
| [05:54] <kenvandine> but finding a fix... has me stumped | |
| [05:54] <kenvandine> RAOF, a fix for that would earn you a few beers at UDS :) | |
| [05:54] <RAOF> Seriously, I'd try throwing a dbus_threads_init_default() early in gwibber's main. | |
| [05:55] <RAOF> That does seem to have fixed a bunch of colord problems. (Possibly exposing more ☺) | |
| [05:55] <RAOF> It also fixed everything using gconf. | |
| [06:02] <kenvandine> dbus.mainloop.glib.threads_init | |
| [06:02] <kenvandine> RAOF, ^^ | |
| [06:03] <RAOF> kenvandine: Ah, it already does? | |
| [06:03] <kenvandine> no | |
| [06:03] * RAOF isn't familiar with what that's hooked up to. | |
| [06:03] <kenvandine> i don't think it does | |
| [06:04] <RAOF> Yeah, looks like that's what you want to call. | |
| [06:05] <kenvandine> i wish the crash was more predictable... | |
| [06:05] <RAOF> Hurray for thread madness! | |
| [06:05] <kenvandine> i am trying it | |
| [06:09] <kenvandine> didrocks, hummm... when i unplug i am getting the fallback icon for indicator-power | |
| [06:09] <kenvandine> perhaps something in that ubuntu-mono upload? | |
| [06:09] <didrocks> kenvandine: right, I noticed that yesterday as well, waiting for larsu to connect :) | |
| [06:09] <kenvandine> ok | |
| [06:09] <kenvandine> cool | |
| [06:09] <kenvandine> brb | |
| === tkamppeter_ is now known as tkamppeter | |
| [06:37] <kenvandine> RAOF, that really might have fixed it! | |
| [06:38] <RAOF> GDBus is a hateful blight upon the world :( | |
| [06:38] <kenvandine> RAOF, you are my hero | |
| [06:39] <RAOF> Which is annoying, because if you use *only* gdbus it's quite nice. | |
| [06:39] <RAOF> It's just that it'll silently break your code if you happen to interact with dbus in any other way :( | |
| [07:07] <kenvandine> ok, i am going to get some sleep | |
| [07:07] <kenvandine> see you guys on monday! | |
| [07:12] <RAOF> desrt: Can I bitch at you about gdbus having a critical design bug causing glib to break everything that uses dbus? | |
| [07:13] <didrocks> desrt: then, it would be my turn with a dconf-writer race issue | |
| [07:13] <didrocks> poor desrt :) | |
| [07:14] <dpm> good morning all | |
| [07:16] <didrocks> hey dpm | |
| [07:17] <dpm> bonjour didrocks | |
| [07:24] <dpm> hey, does anyone know where the strings from the power indicator come from? I see "Battery (0:39 to charge)" shown untranslated on my system, even though the strings (if they come from indicator-power) have been translated for a few cycles. It seems that indicator-power has stopped loading those translations? | |
| [07:31] <didrocks> dpm: I confirm that the only place where this string is exposed for translation is indicator-power on my system | |
| [07:31] <didrocks> dpm: and that it's translated in the .mo file in french as well | |
| [07:31] <didrocks> but the english part is loaded | |
| [07:31] <didrocks> larsu: hey do you know if you init the gettext support? ^ | |
| [07:32] <dpm> thanks didrocks | |
| [07:32] <dpm> seems like a regression | |
| [07:32] <larsu> didrocks, in indicator-power? Let me checl | |
| [07:32] <didrocks> yep | |
| [07:32] <larsu> check | |
| [07:32] <didrocks> larsu: "Battery (charged)" is in indicator-power.mo | |
| [07:33] <didrocks> msgid "%s (charged)" | |
| [07:33] <didrocks> msgstr "%s (chargée)" | |
| [07:33] <didrocks> for instance | |
| [07:33] <didrocks> but showing in english | |
| [07:33] <didrocks> and there is no other strings in other projects exposing it | |
| [07:34] <didrocks> larsu: I have a second request then, the icon isn't right if I unplug the cable (like battery discharging) | |
| [07:34] <didrocks> larsu: still on the unity side, not sure if you can give it a look, should be a mismatching symlink | |
| [07:34] <larsu> didrocks, indicator-power doesn't have a separate service, only the plugin that gets loaded in the panel | |
| [07:35] <larsu> that probably messes with translations, right? | |
| [07:37] <larsu> didrocks, unity-panel-service doesn't intialize gettext, but I guess it would be the wrong domain anyway | |
| [07:37] <didrocks> larsu: yeah, I think the gettext domain is different | |
| [07:37] <didrocks> how does this work for the others? indicator-power is the only to be in that case? | |
| [07:37] <didrocks> (no external service) | |
| [07:37] <larsu> didrocks, I think so, yes | |
| [07:38] <didrocks> is it possible for the plugin to change the gettext domain before loading each strings? | |
| [07:38] <didrocks> and restore afterwards | |
| [07:38] <larsu> tbh, I don't know | |
| [07:38] <tsdgeos> didrocks: larsu: so you guys merged https://code.launchpad.net/~aacid/gtk/ubuntugtk3_nmenomic_patch/+merge/125169 that fixes https://bugs.launchpad.net/ubuntu/+source/nautilus/+bug/896740 How shall i update the bug? | |
| [07:38] <ubot2> Launchpad bug 896740 in appmenu-gtk "With appmenus context-menu shortcut options in nautilus do not work properly. E.g. I need to press R twice to select Rename and then hit Return to select it." [High,In progress] | |
| [07:38] <didrocks> larsu: 37 strings | |
| [07:39] <tsdgeos> mark all those as Invalid and add the "gtk" project and set as fix released there? | |
| [07:39] <larsu> didrocks, where? In i-power? | |
| [07:39] <didrocks> tsdgeos: exactly, thanks! | |
| [07:39] <didrocks> larsu: yep | |
| [07:39] <tsdgeos> didrocks: oki | |
| [07:40] <didrocks> tsdgeos: you can tell it's fixed in 3.5.18-0ubuntu2 | |
| [07:40] <larsu> didrocks, I wonder why this wasn't a problem before... | |
| [07:40] <didrocks> larsu: TBH, I think it was, it's not the first time I'm noticing that | |
| [07:40] <didrocks> larsu: let me start my netbook on precise | |
| [07:41] <didrocks> if I can find my cable power | |
| [07:42] <larsu> didrocks, the real solution here is to move the logic into a separate service, which we are planning on doing anyway next cycle | |
| [07:42] <tsdgeos> didrocks: any hint of the exact name i shall use? gtk? gtk3? can't really find it :_/ | |
| [07:43] <larsu> didrocks, but that would be too much change this late in the cycle. I think nothing else in the panel is using gettext, so we might get away with setting the domain of indicator-power onto the whole panel | |
| [07:43] <larsu> slightly hackish, but.... | |
| [07:44] <didrocks> larsu: sweet, I can't get anymore unity3d running on my precise netbook | |
| [07:44] <didrocks> larsu: however, unity-2d has it translated | |
| [07:44] <didrocks> larsu: as it's using the service, it seems that something changed, right? | |
| [07:44] <didrocks> tsdgeos: gtk+3.0 | |
| [07:44] <larsu> didrocks, charles changed the implementation quite a bit, but I think it was always without a separate process | |
| [07:44] <larsu> didrocks, do you have a indicator-power process running? | |
| [07:45] <larsu> (on precise) | |
| [07:45] <didrocks> larsu: let me look after unf**** my session :) | |
| [07:45] <larsu> haha | |
| [07:45] <tsdgeos> didrocks: and now more questions (i'm on newbiew mode) i got the "precise" subthingie in gtk+3.0, how do i add the Quantal one | |
| [07:45] <tsdgeos> ? | |
| [07:46] <didrocks> larsu: yep, no service | |
| [07:46] <didrocks> tsdgeos: no need to add the Quantal one as it's fixed | |
| [07:46] <didrocks> tsdgeos: the main task is Quantal | |
| [07:47] <larsu> tsdgeos, but for the record: "nominate for series" | |
| [07:47] <tsdgeos> ok i see | |
| [07:47] <didrocks> it's only used to track bugs we want to track | |
| [07:47] <didrocks> and we don't track them once fixed :) | |
| [07:47] <didrocks> larsu: not everyone can do that, you need special launchpad powers :) | |
| [07:47] <didrocks> even for nominating now | |
| [07:47] <didrocks> and you need to be on the downstream task | |
| [07:48] <larsu> oh yeah right, I forgot that I got moved into that special group recently ;) | |
| [07:48] <didrocks> :) | |
| [07:48] <tsdgeos> well, I left https://bugs.launchpad.net/ubuntu/+source/gtk+3.0/+bug/896740 as tidy as i could, appreciate if you could make sure i did it correctly | |
| [07:48] <ubot2> Launchpad bug 896740 in gtk+3.0 "With appmenus context-menu shortcut options in nautilus do not work properly. E.g. I need to press R twice to select Rename and then hit Return to select it." [Undecided,New] | |
| [07:48] <didrocks> larsu: you are special! | |
| [07:48] <larsu> \o | |
| [07:49] <didrocks> tsdgeos: put the main gtk+3.0 task as fix released | |
| [07:49] <didrocks> otherwise, looks good | |
| [07:49] <didrocks> larsu: so something changed, maybe in the unity panel service? | |
| [07:49] <tsdgeos> didrocks: ok, done | |
| [07:50] <didrocks> tsdgeos: looking good, thanks! | |
| [07:50] <larsu> didrocks, I don't think so, that hasn't been touched a lot recently. I'm digging in the precise sources of i-power right now, let's see what I can find | |
| [07:50] <larsu> didrocks, otherwise, charles will probably be able to help us (when he's awake ;) ) | |
| [07:51] <didrocks> larsu: yep :) | |
| [07:51] <didrocks> larsu: so that + the icon (the second one is probably more trivial) | |
| [07:52] <larsu> didrocks, maybe I messed up the icon renaming, I'll have a look right after this | |
| [07:52] <larsu> didrocks, got it! | |
| [07:53] <larsu> didrocks, the old implementation called g_dgettext, which you can pass the domain directly | |
| [07:53] <larsu> the new implementation uses _(), which uses the default domain | |
| [07:54] <larsu> we need more developers using different languages | |
| [07:54] * larsu should switch to german.... | |
| [07:55] <larsu> didrocks, you'll get a fix today | |
| [07:55] <didrocks> larsu: excellent, thanks! | |
| [07:55] <didrocks> dpm: ^ | |
| [07:56] <didrocks> larsu: yeah, please use your desktop in your language! don't hate it :) | |
| [07:56] <didrocks> dpm: did you get a bug for it? | |
| [07:59] <dpm> didrocks, I haven't filed one, as I first wanted to confirm it. Shall I just file a bug? | |
| [08:00] <dpm> larsu, didrocks, our ever amazing translations community to the rescue: <kelemengabor> it was split out into a new file: http://bazaar.launchpad.net/~ubuntu-branches/ubuntu/quantal/indicator-power/quantal/view/head:/src/device.c | |
| [08:00] <dpm> and it includes glib/gi18n.h, instead of glib/gi18n-lib.h | |
| [08:01] <didrocks> using dgettext(), that's what larsu changed | |
| [08:01] <didrocks> larsu: maybe a comment on the first use so that the change doesn't regress? | |
| [08:01] <didrocks> telling it's loaded in process of the unity-panel-service | |
| [08:01] <larsu> didrocks, yup | |
| [08:05] <larsu> didrocks, actually, dpm's fix is the same as mine with much less work: gi18n-lib redefines _() to call g_dgettext | |
| [08:05] <Laney> pitti: thanks! mbiebl finished it up last night, was going to sync this morning | |
| [08:07] <didrocks> Laney: excellent then :) | |
| [08:07] <didrocks> I think we did that in another package | |
| [08:07] <didrocks> a couple of cycles ago | |
| [08:07] <didrocks> I don't remember which one TBH | |
| [08:07] <larsu> didrocks, do you want a new release for that or can you pull the one revision in the package? | |
| [08:08] <didrocks> larsu: I'll gladely steal the commit :) | |
| [08:08] <larsu> didrocks, cool, I'll test and upload in a bit, no need to wait for charles to review for this tiny thing | |
| [08:09] <didrocks> larsu: right, and I'll confirm it as well | |
| [08:09] <larsu> yep | |
| [08:15] <larsu> didrocks, argh, what do I need to install to make `LANG=de_DE ./program` work? | |
| [08:16] <larsu> I installed german from the "Language Support" thing in g-c-c | |
| [08:21] <didrocks> larsu: yeah, the language support through g-c-c should set it to you | |
| [08:23] <larsu> Locale not supported by C library. Using the fallback 'C' locale. | |
| [08:28] <larsu> didrocks, https://code.launchpad.net/~larsu/indicator-power/i18n-lib | |
| [08:28] <larsu> didrocks, I'll push it to trunk when you tell me it works ;) | |
| [08:30] <didrocks> larsu: ok, testing :) | |
| [08:51] <larsu> didrocks, icon fix: https://code.launchpad.net/~larsu/ubuntu-mono/lp1001229/+merge/125430 | |
| [08:52] <larsu> didrocks, you were right, the link target for the battery-full-symbolic were wrong: they were missing the .svg extension | |
| [08:53] <didrocks> larsu: ah! excellent, thanks! | |
| [08:57] <didrocks> larsu: sorry, got other discussions | |
| [09:03] <didrocks> larsu: translation fix confirmed | |
| [09:03] <didrocks> larsu: you can merge it to trunk | |
| [09:03] <larsu> didrocks, will do | |
| [09:03] <larsu> dpm, thanks for pointing me to that other header! | |
| [09:04] <dpm> larsu, np, thanks for the quick fix! | |
| [09:05] <larsu> didrocks, it's in trunk | |
| [09:05] <didrocks> larsu: and it's uploaded :) | |
| [09:05] <didrocks> still waiting on ubuntu-mono to build now | |
| [09:06] <didrocks> larsu: and correct icons (at least, for almost fully charged battery) loaded! | |
| [09:07] <larsu> didrocks, great thanks | |
| [09:08] <didrocks> larsu: thanks to you :) | |
| [09:36] <ronoc> <rant>The "extra stereo" plugin for audious should be deleted, it clearly introduces artifacts into the mix</rant> | |
| === doko_ is now known as doko | |
| [09:42] <doko> pitti, while you are at cairo, there is the cairao-c5 ftbfs (test cases), however the 1.8 package is a mess | |
| [09:42] <doko> see the debian report | |
| [09:48] <pitti> doko: I don't see a package cairo-c5, caiaro-c5, or anything similar? | |
| [09:49] <pitti> (also, I won't get to that today -- beta freeze rush) | |
| [09:50] <doko> pitti, cairo-5c | |
| [10:00] <tsdgeos> guys what's your opinion on https://bugs.launchpad.net/ayatana-scrollbar/+bug/1005677 and the two branches i made to fix it https://code.launchpad.net/~aacid/ayatana-scrollbar/qt_no_overlay_scrollbar/+merge/124697 and https://code.launchpad.net/~aacid/ayatana-scrollbar/blacklisted_libs/+merge/124873 ? I'm getting from Cimi that he is busy and can't review it | |
| [10:00] <ubot2> Launchpad bug 1005677 in ayatana-scrollbar "Re-emergence of "Gtk-CRITICAL **: IA__gtk_widget_style_get: assertion `GTK_IS_WIDGET (widget)'" Makes vlc and other Qt apps crashing crashing" [High,In progress] | |
| [10:01] <tsdgeos> Sincerely i prefer the first one but Cimi says he doesn't like distro patching (and i can agree to that) | |
| [10:09] <pitti> didrocks: oh, nobody packaged gvfs 1.13.8 and .9? I guess I should do that today then | |
| [10:10] <didrocks> pitti: oh? I asked you on Monday, but didn't get any answer, I didn't dare touching it, sorry | |
| [10:10] <pitti> np; we have pretty good test coverage for it now :) | |
| [10:11] <didrocks> pitti: heh, you can upload with your eyes closed :) | |
| [10:11] * didrocks goes back to the PS/unity """party""" | |
| [10:11] <pitti> if I upload to -proposed, yes :) | |
| [10:17] <Mirv> yeah that's where the real party is | |
| === MacSlow is now known as MacSlow|lunch | |
| [11:49] <Laney> larsu: magically, the indicator-power icons are fixed ;-) | |
| [11:49] <larsu> Laney, ;) | |
| [11:57] <xnox> Laney: larsu: thank you. I was scared that it will be default =) | |
| [12:11] <Sweetshark> hmmm, installing the quantal daily CD (amd64) in VirtualBox fails to install the bootloader. known issue? | |
| [12:20] <didrocks> xnox: any idea? ^ | |
| [12:20] <didrocks> Sweetshark: hey! | |
| [12:21] <didrocks> Sweetshark: you have one bug on http://reports.qa.ubuntu.com/reports/rls-mgr/rls-q-tracking-bug-tasks.html, do you think you will have the time to work on it? :) | |
| [12:21] <didrocks> it was targetted for beta2 | |
| [12:22] <xnox> Sweetshark: no idea. Was fine yesterday... but there is reported jenkins failure today (?!) bug bug 1053317 | |
| [12:22] <ubot2> Launchpad bug 1053317 in ubiquity "Failed to symbolic-link /boot/initrd.img-3.5.0-15-generic to initrd.img:File exists at /var/lib/dpkg/info/linux-image-3.5.0-15-generic.postinst line 614. on amd64 installation" [Undecided,New] https://launchpad.net/bugs/1053317 | |
| === MacSlow|lunch is now known as MacSlow | |
| [12:28] <dpm> does anyone know what the indicator-sync is? I've just noticed that new translations template in the queue and approved it | |
| [12:28] <dpm> it seems the template was uploaded yesterday the first time | |
| [12:28] <Sweetshark> didrocks: hmm, like this: http://paste.ubuntu.com/1216635/ ? | |
| [12:29] <didrocks> Sweetshark: the metapackage is in universe? | |
| [12:30] <davidcalle> dpm, it's the indicator that should handle and display things such as Ubuntu One sync progress. U1 should be updated to work with it, which is, I believe, in progress or done. | |
| [12:31] <dpm> davidcalle, ok, thanks. | |
| [12:31] <davidcalle> didrocks, hey | |
| [12:31] <Sweetshark> didrocks: eh, no. how does micahg get that? the libreoffice source package is in main. | |
| [12:32] <didrocks> hey davidcalle! | |
| [12:32] <didrocks> Sweetshark: can you please check with him? | |
| [12:34] <davidcalle> didrocks, there has been a new photo lens upload yesterday, but the packaging has lost a depends (was here in the previous distro version). Neither mterry or kenvandine are around, would you have the time to fix it? (or point me to the good person) https://bugs.launchpad.net/ubuntu/+source/unity-lens-photos/+bug/1053257 | |
| [12:34] <ubot2> Launchpad bug 1053257 in unity-lens-photos "unity-lens-photos crashed with ImportError in /usr/lib/unity-lens-photos/flickr_scope.py: No module named oauthlib.oauth1" [Medium,In progress] | |
| [12:35] <didrocks> davidcalle: ok, will do it | |
| [12:35] <davidcalle> didrocks, thanks :) | |
| [12:36] <didrocks> yw :) | |
| [12:38] <Sweetshark> didrocks: i am asking back on the bug what that is about. | |
| [12:41] <didrocks> ok | |
| [12:43] <didrocks> davidcalle: do you know where the packaging branch is? | |
| [12:43] <didrocks> seems ubuntu/unity-lens-photos | |
| [12:43] <davidcalle> didrocks, don't know, looking for it | |
| [12:43] <didrocks> davidcalle: I guess it's that one | |
| [12:44] <davidcalle> didrocks, yes it is | |
| [12:45] <didrocks> davidcalle: ok, a lot of uploads are missing in this | |
| [12:45] <didrocks> davidcalle: I'm just adding the dep | |
| [12:45] <davidcalle> didrocks, and it misses a depends on python3-oauthlib | |
| [12:46] <davidcalle> didrocks, yes, mterry took over it for some time. | |
| [12:46] <didrocks> without pushing the branch and following the merge-upstream workflow :/ | |
| [12:47] <didrocks> davidcalle: gir1.2-gdata-0.0 dep is needed as well, right? | |
| [12:47] <davidcalle> didrocks, it is, yes. Missing too? | |
| [12:47] <didrocks> yep | |
| [12:47] <davidcalle> :/ | |
| [12:51] <didrocks> davidcalle: uploaded | |
| [12:51] <davidcalle> didrocks, merci! | |
| [12:52] <didrocks> de rien :) | |
| [12:54] <desrt> RAOF: sure. what's up? | |
| [13:32] <cyphermox> didrocks: saw my merge proposal? | |
| [13:33] <didrocks> cyphermox: hum, not really, did you subscribe me to it? | |
| [13:33] <cyphermox> yeah, you're reviewer | |
| [13:33] <cyphermox> it's for unity | |
| [13:33] <didrocks> cyphermox: link? | |
| [13:33] <cyphermox> https://code.launchpad.net/~mathieu-tl/ubuntu/quantal/unity/indicator-sync-recommends/+merge/125342 | |
| [13:34] <didrocks> cyphermox: it's already installed by default? | |
| [13:34] <cyphermox> didrocks: it's not | |
| [13:34] <cyphermox> U1 will need it | |
| [13:34] <didrocks> cyphermox: do we have a FFe? | |
| [13:34] <didrocks> acked by the release team | |
| [13:34] <didrocks> and an UIFe? | |
| [13:34] <cyphermox> it's packaged and there as been a MIR | |
| [13:35] <didrocks> and about the FFe/UIFe? | |
| [13:35] <cyphermox> there is no UIFe/FFe for having it included by Unity; oops | |
| [13:35] <didrocks> I mean, for having it installed by default? | |
| [13:35] <cyphermox> no | |
| [13:35] <didrocks> ok, I won't include it then until that is done | |
| [13:35] <cyphermox> dobey: ^^ | |
| [13:35] <didrocks> cyphermox: it's not that one? https://bugs.launchpad.net/ubuntuone-client/+bug/1042343 | |
| [13:35] <ubot2> Launchpad bug 1042343 in ubuntuone-client/trunk "[FFE] Ubuntu One integration with Q sync indicator" [High,Fix committed] | |
| [13:36] <cyphermox> that sounds more like for U1 | |
| [13:36] <didrocks> 15:34:37 cyphermox | U1 will need it | |
| [13:36] <didrocks> isn't what yo utold? | |
| [13:36] <didrocks> your told* | |
| [13:36] <didrocks> you* | |
| [13:36] <cyphermox> yes | |
| [13:36] <Laney> it's not that | |
| [13:36] <Laney> that is for U1 using the new API | |
| [13:36] <didrocks> Laney: ok :) | |
| [13:36] <Laney> but it is rather pointless to use that if there's nowhere to render it | |
| [13:36] <didrocks> cyphermox: tell me when they will have followed the process, they need an acked FFe and UIFe | |
| [13:37] <dobey> hrmm | |
| [13:37] <Laney> I suppose it will impact screenshots of the desktop | |
| [13:37] <cyphermox> yeah | |
| [13:38] <didrocks> hence the ask for UIFe :) | |
| [13:38] <Laney> indeed | |
| [13:38] <cyphermox> I wonder what happens when there is nothing syncing, hopefully it's passive then | |
| [13:40] <jbicha> and will adding indicator-sync integration mean we can finally kill the U1 shortcut by default in the launcher? | |
| [13:45] <dobey> jbicha: why would we kill the shortcut? i suppose we could at some point, but it doesn't seem really necessary | |
| [13:48] <cyphermox> jbicha: why kill the shortcut? | |
| [13:48] <cyphermox> you need to configure it somehow at first | |
| [13:48] <dobey> and yeah, the sync indicator integration will only work when it's actually configured and running | |
| [13:48] <jbicha> we have way too much stuff in the launcher by default; once configured, I pretty much never use the control panel | |
| [13:48] <dobey> remove the libreoffice shortcuts by default | |
| [13:48] <cyphermox> I don't know how indicator-sync shows things while they aren't syncing, I'm hoping it's PASSIVE, so not showing at all if there is no operation in progress | |
| [13:49] <jbicha> the U1 shortcut is really just an advertisement & not used much so it seemed an easier target for my frustration | |
| [13:50] <cyphermox> hehe | |
| [13:50] <dobey> cyphermox: well if ubuntuone-syncdaemon isn't running (which it isn't by default, as it only runs after you set up u1), and the indicator integration comes from there… | |
| [13:50] <charles> cyphermox: it's in the spec, it shows if there's any client registered on it, whether there's activity or not | |
| [13:50] <jbicha> but sure, we would get rid of Calc & Impress by default | |
| [13:50] <cyphermox> jbicha: OTOH, the control panel is reachable from the... control panel | |
| [13:50] <cyphermox> ykwim ;) | |
| [13:50] <jbicha> which is more logical | |
| [13:50] <dobey> anyway, have to run real quick. brb | |
| [13:51] <cyphermox> bah, it's similar to Software Center. do you use it much? | |
| [13:51] <larsu> charles, which, imo, is the weakest part of the spec. It shouldn't show when it's not syncing | |
| [13:52] <cyphermox> some people might; jsut like some people might use the U1 control center regularly to see how well their transfers are going or if the right directories get synced | |
| [13:52] <jbicha> I have a very difficult time trying to get people to use apt-get, so yes, Software Center is very useful & used more than once | |
| [13:52] <charles> larsu: yeah I was kind of surprised at that ... a lot of the indicator design choices lately have been hide-it-until-you-need-it | |
| [13:52] <cyphermox> charles: oh, so it always does show? | |
| [13:52] <cyphermox> :) | |
| [13:52] <larsu> charles, exactly | |
| [13:52] <charles> cyphermox: it shows if there's any client registered on it, whether there's activity or not | |
| [13:53] <cyphermox> ah, well that's probably not too bad | |
| [13:53] <cyphermox> definitely needs a UIFe/FFe anyway | |
| [13:53] <jbicha> in that case, it sounds like the spec was designed so that we could kill the launcher icon by default | |
| [13:53] <larsu> jbicha, yes! | |
| [13:53] <cyphermox> charles: do you want to file a bug for that if dobey hasn't? Then I'll shepherd it to get the FFe approved | |
| [13:53] <charles> *nod* | |
| [13:54] <charles> jbicha: *nod* | |
| [13:54] <charles> cyphermox: file a bug for not showing until needed? run that by mpt first | |
| [13:54] <cyphermox> no | |
| [13:54] <cyphermox> I mean file a bug for having it installed by default | |
| [13:55] <charles> cyphermox: hm, I thought that already was ticketed. one moment | |
| [13:55] * mpt tries to remember | |
| [13:55] <cyphermox> ok | |
| [13:55] <mpt> Possibly the idea was for it to be an advertisement, removing itself once you were registered? | |
| [13:55] <mpt> (or if you said you weren't interested) | |
| [13:55] <mpt> That's not mentioned in the sync menu spec... | |
| [13:56] <charles> cyphermox: https://bugs.launchpad.net/ubuntu/+source/indicator-sync/+bug/1046055 | |
| [13:56] <larsu> mpt, automatically removing launcher items doesn't sound like a good idea | |
| [13:56] <ubot2> Launchpad bug 1046055 in indicator-sync "[MIR] indicator-sync" [Undecided,Fix released] | |
| [13:56] <cyphermox> charles: that's just to get it in Main | |
| [13:57] <charles> cyphermox: definitely then FFe | |
| [13:57] <cyphermox> hmm.. perhaps my leak fix for nm-applet was overengineered :) | |
| [13:58] <mpt> aha, here we go: https://wiki.ubuntu.com/MenuBar#system-status | |
| [13:58] <mpt> "(Present until you've signed up)" | |
| [13:58] <mpt> That was just a suggestion, though, I'm not an Ubuntu One designer | |
| [14:00] <cyphermox> desrt: glib question for you; how bad is it to use g_slist_append? punishable by death since it needs to iterate through the list? my alternative involves some changes I'm a little worried about | |
| [14:01] <larsu> cyphermox, a common convention is to use g_slist_prepend to add all items, and then call g_slist_reverse | |
| [14:02] <cyphermox> larsu: I want to touch it as little as possible, it's a list I get from an association on a menu item | |
| [14:03] <larsu> cyphermox, I don't understand your question then. Do you want to add a single item to the list? | |
| [14:03] <cyphermox> basically, http://paste.ubuntu.com/1216725/ ; starts at line 238; then 315. | |
| [14:04] <cyphermox> larsu there could be any number of items added to the list | |
| [14:04] <cyphermox> it's a list of duplicate BSS for a given SSID | |
| [14:05] <larsu> cyphermox, it's already using g_slist_prepend, though | |
| [14:05] <cyphermox> (clear_dupes_list frees the whole thing) | |
| [14:05] <cyphermox> larsu: yes | |
| [14:05] <desrt> cyphermox: there's nothing wrong with it perse | |
| [14:05] <charles> cyphermox: I don't see what you're asking about? | |
| [14:05] <cyphermox> larsu: the original (the minuses at 315, appears to leak, according to valgrind | |
| [14:05] <charles> cyphermox: I don't see much list overhead in the patch here | |
| [14:06] <cyphermox> ie. just prepending doesn't change the pointer in the object to the new start of the list | |
| [14:07] <larsu> cyphermox, yeah calling g_object_set_data_full is correct, but why are you copying the list? | |
| [14:07] <larsu> cyphermox, ah! there's a destroynotify set | |
| [14:08] <cyphermox> larsu: because yes | |
| [14:08] <cyphermox> the new | |
| [14:08] <charles> lol | |
| [14:08] <cyphermox> argh | |
| [14:08] <desrt> steal it! | |
| [14:08] <cyphermox> the new g_object_data_full will call the destroynotify | |
| [14:08] <desrt> g_object_steal_data | |
| [14:08] <cyphermox> desrt: won't the above still apply? | |
| [14:08] <charles> g_object_steal_data | |
| [14:08] <desrt> it gets the original pointer and doesn't call the destroynotify | |
| [14:08] <larsu> cyphermox, what charles said | |
| [14:08] <desrt> then you can modify and reset it | |
| [14:09] <larsu> oh, and desrt | |
| [14:09] <desrt> hi larsu :) | |
| [14:09] <larsu> desrt, hi! | |
| [14:09] <cyphermox> desrt: I thought the notify was being called when the set_data was being done | |
| [14:09] <desrt> it is | |
| [14:09] <desrt> but it's called on the _last_ value | |
| [14:09] <cyphermox> oh | |
| [14:10] <desrt> ie: if i call set_full (obj, "x", y, z); | |
| [14:10] <cyphermox> which no longer exists | |
| [14:10] <desrt> then z(y) is called on the *next* call to set() | |
| [14:10] <desrt> if i steal() it first, then 'y' is no longer owned by obj, and z is never called | |
| [14:10] <cyphermox> ah, not just called with something that doesn't exists, not called at all? | |
| [14:11] <cyphermox> this will make things simpler, yeah | |
| [14:11] <micahg> Sweetshark: the libreoffice binary (metapackage) package is in universe, that's what's being prompted about on upgrade | |
| [14:14] <cyphermox> desrt: larsu: charles: thanks. | |
| === m_conley_away is now known as m_conley | |
| [14:19] <Ursinha> bug 1052754, that was fast! | |
| [14:19] <ubot2> Launchpad bug 1052754 in apport "apport-gtk crashed with AttributeError in _filter_tag_names(): 'int' object has no attribute 'isspace'" [Critical,Fix released] https://launchpad.net/bugs/1052754 | |
| [14:19] <Ursinha> thanks pitti :) | |
| [14:19] <pitti> well, it was me breaking it in the first place :) | |
| [14:20] <Sweetshark> micahg: we are splitting one source package in multiple reporsitories? | |
| [14:20] <micahg> Sweetshark: yes, it's common | |
| [14:20] <micahg> well | |
| [14:20] <micahg> not exactly | |
| [14:21] <micahg> the binary is in universe, the source in main | |
| [14:22] <micahg> sometimes this is done where binary dependencies are needed for something that can't/shouldn't go through the MIR process | |
| [14:23] <micahg> Sweetshark: the problem is it's confusing for the user, so the short description just needs to be tweaked so it's clear that it's only the metapackage being demoted, not libreoffice itself (or promote the metapackage back to main if there's no reason for it to be in universe) | |
| [14:23] <Sweetshark> micahg: well, libreoffice source is obviously MIRed (otherwise it couldnt produce main binaries), so it doesnt make sense at all to have it put stuff in universe. | |
| [14:24] <micahg> Sweetshark: not necessarily true, we have many packages in main with binaries in universe | |
| [14:26] <Sweetshark> 14:28 < Sweetshark> didrocks: hmm, like this: http://paste.ubuntu.com/1216635/ ? | |
| [14:26] <Sweetshark> ^^ micahg | |
| [14:27] <micahg> something like that | |
| [14:28] <didrocks> I think most of people are installing "libroffice" | |
| [14:28] <didrocks> so libroffice in main makes more sense | |
| [14:29] <micahg> yes, if it's not pulling a heap of other stuff in main (unless that stuff is supportable) | |
| [14:30] <didrocks> Sweetshark: can you check that? if so, I'll do the promotion | |
| [14:31] <micahg> didrocks: you'll need to add it to the supported seed as well or it'll fall right back out of main | |
| [14:31] <Sweetshark> micahg: k, will do that (its trivial). OTOH this unclean separation still boggles my mind. see didrocks comments, mind that we have packagekit integration in libreoffice now that might have interesting implications and he fact that an source build of a main package geenerates nonmain binaries is really icky too me. | |
| [14:32] <dobey> mpt: i don't think we'd remove the u1 launcher automatically after sign-up. that would be quite odd to have happen. and we also do integrate with the unity launcher APIs already as well | |
| [14:32] <micahg> Sweetshark: main/universe distinction will hopefully go away by the next LTS, but the idea is that not everything is Canonical supported | |
| [14:33] <didrocks> micahg: thanks, I'm an archive admin, I know about this :) | |
| [14:33] <micahg> so, if a binary is not needed by the other stuff in main and pulls in a bunch of other things also not needed, it's demoted (if possible) | |
| [14:33] <didrocks> and a core dev for years | |
| [14:33] <micahg> didrocks: figured as much, just wanted to be sure :) | |
| [14:34] <Sweetshark> micahg: I understand that. but since the source already needs the build-deps that only applies to stuff with nasty binary-only closed source deps. right? | |
| [14:35] <micahg> well, sometimes there's a run time dependency not needed at build time | |
| [14:38] <micahg> Sweetshark: didrocks: the libreoffice binary as is doesn't have its dependencies in main (libreoffice-filter-mobiledev, ttf-sil-gentium-basic) | |
| [14:38] <didrocks> micahg: ah, so makes sense it's been demoted | |
| [14:38] <didrocks> Sweetshark: are those mandatory, can we put them as suggest? ^ | |
| [14:40] <Sweetshark> didrocks: can be demoted to suggest easily. | |
| [14:40] <didrocks> Sweetshark: can you please do so rather than the description? | |
| [14:40] <mpt> dobey, that something has APIs doesn't necessarily mean that it's a good idea to be part of it. :-) | |
| [14:40] <micahg> Sweetshark: didrocks: well, the description should be updated as well (metapackages should be labeled as such) | |
| [14:41] <mpt> dobey, I agree going away after sign-up would be unusual, but maybe it's a choice between doing that, and not being in the Launcher at all. | |
| [14:41] <micahg> but the severity is lower if it's in main | |
| [14:41] <didrocks> micahg: is it mandatory but the debian policy? never seen this paragraph | |
| [14:45] <micahg> didrocks: well, not exactly, this quote from 3.4.1 seems most relevant: "Remember that in many situations the user may only see the synopsis line - make it as informative as you can. " | |
| [14:47] <didrocks> micahg: I have nothing against adding it if it doesn't diverge from debian and/or we don't have translations for it | |
| [14:48] <micahg> didrocks: agreed, it could certainly wait until next cycle at this point and should go through Debian if the binary is promoted | |
| [14:48] <didrocks> yep | |
| [15:00] <Laney> pitti: hey, is it OK for me to upload for #1042231? | |
| [15:01] <pitti> Laney: I guess so, but I guess mvo or glatzor would be better to ping | |
| [15:02] <pitti> Laney: but I guess with the beta freeze, it's better to fix it now | |
| [15:02] <Laney> ah, sorry, thought you were the distro guy for that | |
| [15:02] <tedg> mterry, tsdgeos implemented saving the last server, but it uses a new lib as it encrypts the cache. Does that need an FFe? | |
| [15:03] <tedg> mterry, For RLS | |
| [15:03] <mterry> tedg, guh, yeah. My understanding is all new packages need FFe | |
| [15:03] <tedg> mterry, Not a new package, a new build dep. | |
| [15:04] <tedg> mterry, The dep is already in main. | |
| [15:04] * tedg forgot to verify that before asking, but did now. | |
| [15:04] <mterry> tedg, ah, few. no, shouldn't need an FFe | |
| [15:05] <mterry> I mean phew | |
| [15:05] <tedg> Hooked on phonics strikes again! | |
| [15:05] <tedg> :-) | |
| [15:05] <tedg> mterry, Ah, okay. I'll pull it in. | |
| [15:05] <tedg> mterry, dbarth helped add some testing to libpam-freerdp as well. So I'll be releasing that. | |
| [15:07] <mvo> Laney: go for it | |
| [15:08] <Laney> cheers | |
| [15:14] <dobey> charles, cyphermox, didrocks: is there an FFe bug filed for indicator-sync to be installed by default? | |
| [15:14] <didrocks> dobey: I think the question was addressed to your team about that | |
| [15:14] <didrocks> dobey: I never saw that | |
| [15:14] <didrocks> and as told, there is the need to an UIFe and a FFe | |
| [15:15] <dobey> right; i saw some backlog with charles and cyphermox discussing that, but don't see a bug for it. | |
| [15:16] <dobey> the MIR bug was linked, but that's fixed and separate | |
| [15:17] <Laney> mvo: could you merge lp:~laney/sessioninstaller/fix-1042231 into the packaging branch please? | |
| [15:17] <mterry> jdstrand, you seemed semi-involved in the various webapps FFe bugs. Have you done any security auditing of those packages? I'm starting a pre-review MIR for them | |
| [15:19] <mvo> Laney: sure, hm, it says bzr: ERROR: Not a branch: "bzr+ssh://bazaar.launchpad.net/~laney/sessioninstaller/". | |
| [15:19] <Laney> er | |
| [15:20] <Laney> oh, sorry | |
| [15:20] <Laney> lp:~laney/sessioninstaller/distro-fix-1042231 | |
| [15:20] <dobey> cyphermox, charles: so does that uife/ffe bug still need to get filed? | |
| [15:21] <Laney> mvo: ^ | |
| [15:25] <mvo> Laney: done, thanks! | |
| [15:25] <Laney> cheers | |
| [15:29] <didrocks> dobey: they need to. Not sure they can be reviewed today TBH | |
| [15:29] <didrocks> dobey: it's really late, we are preparing the same thing with PS, but started 2 weeks ago to discuss the exceptions | |
| [15:29] <didrocks> and still in flux :) | |
| [15:30] * didrocks needs to reboot, brb | |
| [15:32] <robru> hey didrocks | |
| [15:32] <didrocks> hey robru | |
| [15:33] <robru> that rygel ftbfs has me quite stumped. I tried a git bisect on it and was unable to find a version that actually does build successfully, which leads me to believe that perhaps something changed in some package that rygel depends on, which broke it, rather than a change in rygel itself. | |
| [15:33] <robru> but I don't know enough vala to really say what. The test that fails is related to Soup though. | |
| [15:35] <didrocks> robru: ok no worry, do you want something else to work on? | |
| [15:35] <didrocks> or do you have enough on your plate? | |
| [15:35] <robru> didrocks: I have quite a lot of gwibber work left ;-) | |
| [15:36] <robru> but maybe if you want to throw me something easy ;-) | |
| [15:36] <didrocks> robru: ok, go for it then :) | |
| [15:36] <robru> thanks | |
| [15:36] <jdstrand> mterry: I have not. mdeslaur looked at 1043461. chriscoulson looked at the extension code and I trust his judgement. the bit we need to make sure of is that there is a process in place regarding testing before our security updates for the browser go out. we can't wait on them for updates, but we recognize breaking webapps is bad | |
| [15:36] <didrocks> robru: quite busy, will try to find an easy one for you tomorrow :) | |
| [15:36] <robru> didrocks: ok thanks | |
| [15:36] <didrocks> yw :) | |
| [15:36] <mterry> jdstrand, webapps-applications depends on gjs (which depends on mozjs). Are those giant red flags? | |
| [15:37] <jdstrand> probably, but it isn't like we can block it. chris looked at this stuff, and I know mdeslaur has also | |
| [15:37] <mterry> jdstrand, fyi, MIR bug I started is bug 1053463 | |
| [15:37] <ubot2> Launchpad bug 1053463 in ubuntu "[MIR] webapps packages" [Undecided,New] https://launchpad.net/bugs/1053463 | |
| [15:37] <mterry> jdstrand, ok | |
| [15:38] <jdstrand> (well, mdeslaur has looked at this stuff at a high level-- I don't know if he specifically looked at webapps-applications) | |
| [15:38] * jdstrand guesses not | |
| [15:38] <mdeslaur> no, I just looked at how it was installing packages | |
| [15:38] <mdeslaur> I didn't take a close look at the code | |
| [15:38] * jdstrand nods | |
| [15:39] <mterry> jdstrand, mdeslaur: which of you should I ask to review things as they come up? (I can do MIR side, but for auditing purposes) | |
| [15:39] <mterry> Or shall I just assign to team | |
| [15:41] <jdstrand> mterry: the team is fine | |
| [15:41] * mterry suspects most of this stuff will need a security pass | |
| [15:42] <jdstrand> mterry: how much are we talking about? this is extremely late for any sort of timely review | |
| [15:42] <jdstrand> (for beta2) | |
| [15:42] <dobey> cyphermox, charles, didrocks: https://bugs.launchpad.net/ubuntu/+source/indicator-sync/+bug/1053482 | |
| [15:42] <ubot2> Launchpad bug 1053482 in indicator-sync "[FFe] [UIFe] Install indicator-sync by default" [Undecided,New] | |
| [15:43] <mterry> jdstrand, tell me about it :) Um, well, chris can probably verify the quality of the FF extensions. So I guess just gjs and mozjs? Unless Chris can vouch for mozjs too? | |
| [15:43] <dobey> does that need to be on unity? or affects both? | |
| [15:46] <didrocks> dobey: affects unity and indicator-sync | |
| [15:46] <didrocks> dobey: then, you need to get the release team and the doc team to ack it | |
| [15:47] <didrocks> dobey: TBH, I'm not available today, still continuing the PS and unity work | |
| [15:49] <jbicha> olli: I'm curious about the UI for bug 1053470 You said it's a lens, but it only shows an additional category after searching | |
| [15:49] <ubot2> Launchpad bug 1053470 in unity "[FFE][UIFe] Dash - land the "more suggestions..." lens" [Critical,Fix released] https://launchpad.net/bugs/1053470 | |
| [15:50] <jbicha> does that mean it doesn't show up in the white lens bar at the bottom of the dash? | |
| [15:50] <olli> jbicha, it is an additional category | |
| [15:51] <jbicha> so it's more of a scope than a lens? | |
| [15:51] <davidcalle> jbicha, it's a property of lenses that can be set : appear in Dash bar/don't | |
| [15:52] <jbicha> oh, we don't have any other lenses like that yet, right? | |
| [15:52] <mterry> chrisccoulson, how do you feel about mozjs and gjs finding their way into main? | |
| [15:53] <davidcalle> jbicha, not by default, but the 3rd party lens that brings a calculator and weather in the Home dash, does the same thing. | |
| [15:54] <njpatel> jbicha: hey | |
| [15:54] <njpatel> you had a question about the shopping lens not showing up at the bottom? | |
| [15:55] <njpatel> You're right in that it just shows up once you've done a search in the home screen (and if something matches) | |
| [15:56] <njpatel> There isn't anything else that does that right now but it's been a feature since 10.10 of the lenses API | |
| [15:57] <jbicha> njpatel: ok, I'm just trying to get a better understanding of these terms, scopes & lenses are tricky to try to explain, & I guess this is one more facet to understand | |
| [15:57] <njpatel> jbicha: yeah, main thing is only lenses are allowed to show up in the home screen | |
| [15:58] <njpatel> so if you want to be there, but not be a full-on lens with an icon etc | |
| [15:58] <njpatel> then you can | |
| [15:58] <njpatel> but as a scope,your only method of having results on the homescreen is if you are adding results to a lens | |
| [15:59] <jbicha> it sounds like it's a super-scope to me :) | |
| [15:59] <didrocks> jdstrand: unity-lens-shopping in NEW (quantal-proposed) | |
| [15:59] <njpatel> jbicha: hehe | |
| [15:59] * njpatel needs better naming | |
| [15:59] <jbicha> I mean it's like a scope but it can plug into lots of lenses | |
| [16:00] <njpatel> jbicha: actually, something like that is proposed for 13.04 | |
| [16:00] <njpatel> :) | |
| [16:00] <mdeslaur> mterry: seeing as how mozjs is ripe with known security vulnerabilities because the standalone version is unmaintained, I'd say it has a pretty slim chance of making it in main | |
| [16:00] <jbicha> well that's how it feels to a user at least | |
| [16:00] <mterry> mdeslaur, ah OK. That's what I feared | |
| [16:00] <Laney> hm, there's no Filter Results on the home screen | |
| [16:00] <njpatel> which should simplify all these daemons etc | |
| [16:00] <njpatel> Laney: nope | |
| [16:00] <mdeslaur> mterry: probably worth opening a separate MIR for it | |
| [16:00] <njpatel> Laney: that's meant to be like that | |
| [16:01] <Laney> I'm sure that's the case | |
| [16:01] <mterry> mdeslaur, is seed any better from a security point of view? | |
| [16:01] <Laney> I was just assuming you'd be able to turn shopping on and off there | |
| [16:01] <njpatel> Laney: right, the way we say is through software center | |
| [16:01] <njpatel> it would have been nice to have a disable and not remove | |
| [16:01] <njpatel> you're right | |
| [16:05] <g0twig> please checkout the bug here: https://bugs.launchpad.net/ubuntu/+source/gnome-control-center/+bug/871985/comments/17 and this comment | |
| [16:05] <ubot2> Launchpad bug 871985 in gnome-control-center "The printing panel shouldn't display errors about firewalld not running" [Medium,Confirmed] | |
| [16:11] <mdeslaur> mterry: well, seed is based on webkit, which is problematic also | |
| [16:12] <mterry> but at least in main already. :) | |
| [16:13] <mdeslaur> mterry: not for long | |
| [16:13] <mterry> fair enough | |
| [16:13] <mdeslaur> mterry: we have a strategy for webkit, jdstrand can tell you more about it | |
| [16:13] <mterry> interestingly, mozjs is only used for building | |
| [16:13] <mterry> so that's somewhat good. but it would still involve having it in main | |
| [16:17] <mdeslaur> for the build? why does it need a js interpreter for building? | |
| [16:17] <mdeslaur> ah, for the test suite | |
| [16:59] <mterry> chrisccoulson, ping about webapps plugins | |
| [19:45] <jbicha> bkerensa: your system-config-printer fix should be an actual patch so that we can easily apply it to new versions, do you have time to fix? | |
| [19:48] <bkerensa> jbicha: surely | |
| === om26er_ is now known as om26er | |
| [20:58] <jbicha> bkerensa: do you have that patch? apparently we just froze now but I should be able to still upload it | |
| === m_conley is now known as m_conley_away | |
| [21:18] <cyphermox> jbicha: that being for the firewalld check in control-center/ | |
| [21:19] <cyphermox> ? | |
| [21:45] <jbicha> cyphermox: no, bug 1053189 | |
| [21:45] <ubot2> Launchpad bug 1053189 in system-config-printer "UIFE: Rename "Print Settings" to "Printers"" [Low,Triaged] https://launchpad.net/bugs/1053189 | |
| [21:47] <cyphermox> ah | |
| [21:53] <jbicha> the package doesn't use quilt :( | |
| [22:06] <jbicha> cyphermox: hey, it looks like s-c-p isn't in the desktop set, do you want to sponsor for me? | |
| [22:06] <cyphermox> jbicha: you're funny | |
| [22:07] <jbicha> I try | |
| [22:08] <cyphermox> hmm.. sorry; I'd rather not upload things at this point past the freeze deadline | |
| [22:08] <jbicha> or I could ping robert_ancell if you're done for the day | |
| [22:09] <cyphermox> I mean, system-config-printer probably builds fast, but it's in the default install, the rule is you're supposed to avoid it if possible and this doesn't fix a critical issue ;) | |
| [22:10] <jbicha> it affects translations so if we're going to change it, we need to do it now | |
| [22:10] <Laney> you can upload it into the queue, it's fine | |
| [22:10] <jbicha> Laney: LP killed that since it's main, but not the desktop set | |
| [22:10] <jbicha> http://people.ubuntu.com/~jbicha/system-config-printer/ | |
| [22:11] <Laney> not you, people in general | |
| [22:11] <Laney> I'm going to bed, but I'll do it tomorrow if needed | |
| [22:11] <Laney> rather not though, because then I can't do the queue review | |
| [22:11] <stgraber> cyphermox: can you please upload that so we can review and let it in? | |
| [22:12] <jbicha> I somehow had it in my head that the freeze was at midnight UTC so I wasn't watching the clock close enough | |
| [22:12] <stgraber> freezes are pretty much always at 21:00 UTC | |
| [22:12] <jbicha> yeah I should know better by now | |
| [22:14] <Laney> jbicha: how is that different from the branch on the bug? | |
| [22:14] <cyphermox> stgraber: yeah sure | |
| [22:15] <cyphermox> Laney: I'll upload it | |
| [22:15] <jbicha> Laney: it wasn't formatted as a patch, it just modified the source directly | |
| [22:15] <Laney> I see | |
| [22:15] <Laney> well, personally I'd have preferred a debdiff (or branch) on the bug rather than uploading to a web server | |
| [22:15] <Laney> but /me steps away :P | |
| [22:16] <Laney> night | |
| [22:16] <cyphermox> night | |
| [22:21] <cyphermox> ah, indeed looks a little more consistent with the other items | |
| [22:24] <cyphermox> stgraber: done | |
| [22:24] <cyphermox> jbicha: ^ | |
| [22:24] * cyphermox runs to a LUG meeting | |
| [22:25] <bkerensa> ;d | |
| [22:25] <bkerensa> jbicha: why did you need a patch specifically? | |
| [22:25] <bkerensa> the branch should have worked | |
| [22:25] <cyphermox> bkerensa: was it your first patch? | |
| [22:26] <bkerensa> LOL | |
| [22:26] * bkerensa facepalms | |
| [22:26] <bkerensa> no not by far | |
| [22:26] <bkerensa> it was jbicha's patch | |
| [22:26] <cyphermox> oh ok ;) | |
| [22:26] <bkerensa> I submitted a MP | |
| [22:26] <bkerensa> he wanted a patch for some reason | |
| [22:26] <cyphermox> but I mean, first upload? | |
| [22:26] <bkerensa> no | |
| [22:26] <cyphermox> k | |
| [22:27] <cyphermox> brb | |
| [22:27] <jbicha> it needs to be a patch so that we can separate the source from the packaging, which is important for when we want to include a new upstream version | |
| [22:28] <bkerensa> jbicha: ahh its a debian package? | |
| [22:28] <jbicha> everything is a debian package :) | |
| [22:32] <bkerensa> jbicha: debian maintained | |
| [22:33] <jbicha> for most packages, we only touch the debian/ directory and use patches to manipulate the source when needed | |
| [22:34] <jbicha> for debian source format 3.0 packages, you can't touch the source directly so you need to use quilt patches | |
| [22:35] <thumper> hi folks | |
| [22:35] <micahg> well, 3.0 (quilt) | |
| [22:44] <robert_ancell> RAOF, I did some colord package updating (bug 1053701, bug 1053696) | |
| [22:44] <ubot2> Launchpad bug 1053701 in ubuntu "[needs-packaging] colord-gtk" [Wishlist,Triaged] https://launchpad.net/bugs/1053701 | |
| [22:45] <ubot2> Launchpad bug 1053696 in colord "Update to 0.1.23" [Wishlist,Triaged] https://launchpad.net/bugs/1053696 | |
| [22:45] <RAOF> robert_ancell: Oooh, ta. | |
| [22:50] <RAOF> robert_ancell: I wasn't considering it a priority because there's not much in the new versions, other than splitting out colord-gtk into a new source package. Was there a trigger for your welcome updating spree? :) | |
| [22:50] <robert_ancell> RAOF, just cause it was out of date and I like things to be up to date :) | |
| [22:51] <robert_ancell> I guess it won't get into quantal but it should be good to go for Debian and then it's ready to pick up for R | |
| [22:57] <bkerensa> I wish Management Service could have been renamed for 12.10 | |
| === robru is now known as robru|dinner | |
| [23:37] <bschaefer> hello, so who would I bug to get bamf and libunity-webapps rebuilt in unity-team/staging? | |
| [23:37] <bschaefer> https://bugs.launchpad.net/libunity-webapps/+bug/1053688 | |
| [23:37] <ubot2> Launchpad bug 1053688 in libunity-webapps "bamfdaemon links to libunity_webapps.so but the library is named libunity-webapps.so" [Undecided,New] | |
| [23:37] <bschaefer> because of that bug ^ | |
| [23:37] <bschaefer> causes bamfdaemon to die, which causes a bunch of problems in unity :( | |
| [23:38] <cyphermox> bschaefer: I'd recommend asking in #ubuntu-unity too, in case someone is up/looking there | |
| [23:38] <bschaefer> cyphermox, alright thanks! | |