[05:03] Good morning [06:37] @#$#@!!! Morning. [06:42] good morning [06:43] bonjour didrocks [06:43] BigWhale: what's up? :-) [06:43] salut pitti, ça va? [06:43] ça va bien! et toi? [06:43] * pitti was watching you live yesterday [06:43] ça va, ça va :) [06:43] UbunTv! [06:43] pitti: oh, which session? :) [06:44] integrating the phablet apps into ubuntu [06:44] bonjour á tous [06:44] ah ok :-) [06:44] bonjour highvoltage ;) [06:44] bonjour highvoltage, comment vas-tu? [06:44] bien merci et toi? [06:44] pitti: I find it a little bit difficult to lead a session, with all the tabs opened + the hangout window [06:44] I guess it's a question of habit [06:44] yeah [06:45] following a session with video, irc, and etherpad works rather well I find [06:45] pitti, I'm up, that's what .. :/ six hours of sleep in two days and today will be the same and I have to make a new release of Kazam so it will get uploaded to Raring. [06:45] but I found it way more tiring than leading a session at a non virtual UDS [06:45] right [06:45] but not when having an "active" hangout [06:45] exactly [06:45] not sure what can be done for that though [06:45] BigWhale: trying sleep compression? [06:46] didrocks: not sure either, maybe hangout on the tablet/netbook/second screen? [06:46] with a bigger monitor, having two windows side by side ought to work, too [06:46] pitti: yeah, I think that's a viable option [06:46] that gets quite crammed with my 1280 pixels, though [06:46] Two people should lead the sessions. One taking care of video and the other of IRC. And the second person would then pass the info from irc to the first person [06:47] it's hard to keep track of all the people chatting in irc [06:47] BigWhale: yeah, that doesn't fix the "issue" with sessions where you have a ton of tabs [06:47] yeah, we need some kind of moderation there [06:47] * RAOF will get to see this in action when there's a vUDS that doesn't go from 1am to 6am. [06:47] but agree that someone modering IRC is a first step :) [06:48] moderating* [06:48] pitti, I'll just stuff myself with energy drinks and green smoothies :> [06:52] BigWhale: green smoothies, what do you put in it? :) [06:53] didrocks, lots of stuff that tastes awful and that really wakes you up! ;> [06:53] :) [06:53] everything from cabbage to broccoli [06:54] lettuce [06:54] mint ... and few more stuff that I'd have to google their English names :) [06:54] interesting, I drinked a lot of smoothies in Ireland, but never such ones :) [06:54] ahah [06:55] RAOF: not missing much [06:56] rolling release session was quite good, but inconclusive :) [06:56] ah well, have to run to the office. [06:56] later :) [07:48] morning folks [08:02] hey czajkowski [08:28] hey desktopers [08:38] seb128: ho === Sweetsha1k is now known as Sweetshark [08:38] Sweetshark, hey, how are you? [08:40] Mostly fine, except from the "arrgh fuuu LibreOffice on raring ate my data" mail from jono in the morning. [08:40] oh, it did? [08:42] seb128: we should update away from that beta ASAP. 4.0.1 final is going to be released ... *cough* soon, so we should update along with that. [08:43] I have a 4.0.1~rc2 at https://launchpad.net/~bjoern-michaelsen/+archive/libreoffice-nattytest2 to be copied over to the ppa when its done. [08:43] is there any other blocker since you fixed the libcmis issue? [08:43] rc2 is expected to be final. [08:43] no other blockers except DMB review ;) [08:51] Sweetshark, great, let's hope bdrung has time to review/sponsor it ;-) [08:52] seb128: I see no trouble there, bdrung was very quick with the libcmis thing. [08:53] excellent [09:10] pitti, hey, I saw you assigned you the gvfs bug, do you plan to do the update as well? ;-) [09:11] howdy [09:11] Laney, hey, how are you? [09:12] good, but tired! We went to a comedy show last night which started/finished quite late [09:12] how about you? [09:13] I'm good thanks [09:13] I was too exhausted after vUDS to go anywhere though :p [09:14] yeah I'd have rather it was another day but we already booked it [09:17] I hope the show was good at least ;-) [09:17] yep :) [09:45] * seb128 uninstall a stack of kernels and gets back 1GB of disk space [09:46] * chrisccoulson should try that [09:47] http://paste.ubuntu.com/5589957/ !!!!!!! [09:47] Laney, see! [09:48] Laney, do you have a magical command or did you list them by hand? (I do list by hand) [09:48] using autoremove doesn't clean those for me [09:48] oh, that was just autoremove [09:49] autoremove clearly does for me :) [09:49] just cleaned 2.7 GB, I kept the old kernels for a long time due to the network issue [09:52] autoremove freed up nearly 1GB here, but it left most old kernels [09:52] i guess i'll clean those manually [09:53] perhaps it only notices ones which were installed after that feature got added [09:54] i've got 3.0 kernels installed here. how old are those? [09:54] oneiric ish [09:56] nice, another 3GB [09:56] quick, fill the space with another chromium build! [09:57] how to make chrisccoulson - free him up space [09:57] czajkowski, i literally fight over tens of MB here, so a few GB is quite welcome [09:57] * chrisccoulson wishes for a bigger SSD [09:58] same here [09:58] 80G ssd is short [09:58] yeah, i struggle with 120GB! [09:59] which reminds me. i'm sure that i'm due a laptop refresh ;) [09:59] yeah I'm thinking of investing in a desktop and throwing a lot of storage at it, having to do testing in Vms is driving me insane [09:59] I love this laptop but it's not built for testing [10:00] it does test my patience from time to time [10:00] or lack of... === vrruiz_ is now known as rvr_ [10:02] do you have an SSD? [10:02] made such an amazing difference when I upgraded just that one thing on my laptop [10:02] yeah, same here [10:02] I wonder if I should format it though [10:04] https://plus.google.com/u/0/107564545827215425270/posts/V3UH5SUXJcT [10:05] i've got 4.5GB of free space now \o/ [10:06] chrisccoulson: I have a build tree that would just fit in there nicely .... [10:06] bdrung: http://people.canonical.com/~bjoern/libreoffice4/ <- 4.0.1 proposal btw ... [10:06] Sweetshark, you can get a build tree in 4.5GB? lucky you ;) [10:07] chrisccoulson: a developer build, not a release build. [10:07] ah :) [10:19] Hi. I've created a custom indicator. To find out about the available icons, I used the icon library browser. But now I'm finding out that they don't look the same in unity and cinnamon [10:19] marga, right, cinnamon uses a different icon theme [10:19] For example, I'm using emblem-default-symbolic, and in cinnamon it looks awfully big [10:20] seems like a bug in their theme then? [10:20] :-/ [10:21] Well, I'm not sure... It's the same icon, just bigger [10:21] I think it's the indicator that somehow is screwing this up [10:22] marga, why? [10:22] because the icon is available in multiple sizes, but it's showing the wrong size [10:23] maybe it's not available in the wanted size in the theme? [10:23] Sorry, it's actually a scalable svg [10:23] the icon library browser shows 3 sizes, by scaling [10:24] marga, where did you get the cinnamon icon theme? [10:24] it doesn't seem to be into the archive [10:24] uhm... it's the gnome theme, I think [10:25] the library browser says 'inherited from gnome' [10:25] It's in the gnome-icon-them-symbolic [10:26] seb128: I don't think there's an "update" as such, it requires debugging why it doesn't work [10:26] oh, there's 1.15.4, indeed [10:26] pitti, versions disagrees with you [10:26] pitti, ;-) [10:27] ah, released on Monday, that was after I checked last [10:29] seb128: but still, support for powering off was in 1.15.1, so it's still not just the new version [10:29] pitti, well, it's build time enabled though, did you try rebuilding with the new udisk? [10:29] yes [10:29] ok [10:29] pitti, the "update" comment was orthogonal, I've "update gvfs" in my tomboy [10:30] but that tomboy note is endless, so I'm happy to let it to somebody else :p [10:30] ah /me was looking at that [10:30] happy to leave it to pitti though if you're debugging something else anyway [10:30] Laney, there is a bunch of other GNOME updates to do if you feel like doing those [10:30] Laney: ok, WFM; I'll do the update, and debug the poweroff thing [10:30] I was just going to do gtk3 [10:31] seb128: is there a new 3.6.5? or are we going for 3.7.x at last? [10:32] Laney, gssdp and gupnp (sync the new version from debian?), libpango/glib/gtk-perl, glibmm, gnome-keyring, gtkmm3, libsecret, etc [10:32] * pitti HUDs for "disks", gets totem, o_O, and tries again [10:32] pitti, still baking 3.7 in the ppa [10:33] pitti, we have some regressions still, theming issues [10:33] https://bugzilla.gnome.org/show_bug.cgi?id=694217 [10:33] Gnome bug 694217 in general "drag&drop half-broken – items get stuck while dragging" [Normal,Unconfirmed] [10:33] gnome-screenshots turn your screen white [10:33] meh, and searching for d-feet doesn't work at all -- what the heck is wrong with hud search? [10:33] pitti, hud or dash? [10:33] pitti: yeah, regression in the dash [10:33] hud is looking in menus only [10:33] sorry, dash [10:33] pitti: the fix is merging [10:34] with a test :p [10:34] * pitti hugs didrocks [10:34] regression from the new libcolumbus version [10:34] what didrocks said [10:34] * didrocks hugs pitti back [10:35] marga, I don't really have time to look at the issue atm, it's weird because that symbolic icon is only shipped in one package ... does the cinnamon theme inherit from GNOME? [10:35] it does, yes [10:35] marga, in any case have a bug report with a small testcase would be useful [10:35] larsu can probably help you to figure out what's wrong when he's up in a few hours [10:35] at least calling PowerOff on udisks2's d-bus iface works, so gvfs' fault [10:35] Ok, will look a bit deeper into this to try to find the culprit [10:41] bah, typing "spotif" into the dash gets me gedit instead of spotify [10:42] Laney: sounds like the same error I just complained about, where didrocks says it's being fixed? [10:42] oh, didn't read scrollback, sorry [10:42] ah [10:42] yes, thanks [11:03] seb128, hi [11:07] pitti: so I used the g-doc parser against Gtk.gir to generate the docs. I see all the classes in the dot notation & hierarchy which is nice. But no functions/parameters/signals?! [11:08] cyphermox hi, wrt https://bugs.launchpad.net/ubuntu/+source/network-manager-applet/+bug/965895 [11:08] Launchpad bug 965895 in network-manager-applet "After boot, NetworkManager indicator menu only shows four entries" [High,In progress] [11:10] xnox: signals are in the per-class docs; I also see the arguments of methods and signals [11:10] pitti: hmm... [11:10] xnox: the functions seem to be there, but that part is buggy [11:10] Index.main [11:10] Runs the main loop until Index.main_quit is called. [11:10] that should say "Gtk", not "Index" [11:11] pitti: here is what I have: http://people.canonical.com/~xnox/yelp/doc/index.html [11:11] did I generate it wrong? [11:11] there are .pages as well... [11:11] hm, that looks incomplete indeed; I was checking in ylep [11:11] yelp, too [11:12] g-ir-doc-tool --language=Python -o /home/tdlk/build/doc /usr/share/gir-1.0/Gtk-3.0.gir [11:12] * pitti runs yelp-build [11:12] yelp index.page is same.... [11:12] for me anyway. [11:14] i do have -dev -doc etc packages installed. Should I be doing this doc generation in the source package? [11:14] xnox: hm, in *.html I get the complete docs, too [11:16] xnox: so compare http://people.canonical.com/~xnox/yelp/doc/Gtk.FontChooser.html with http://people.canonical.com/~pitti/tmp/Gtk.FontChooser.html [11:16] xnox: (sorry, didn't copy the .js and css bits along) [11:16] done now [11:16] pitti: nice. So how come i'm not getting that?! =)))) [11:17] xnox: did you not feed it enough cookies? [11:17] no off-hand idea, I'm afraid [11:17] gobject-introspection 1.35.8+git20130220-0ubuntu1 [11:17] xnox: I have yelp-tools and yelp-xsl installed [11:17] libgtk-3-dev 3.6.4-0ubuntu6 [11:18] pitti: yeah, same here. [11:18] xnox: but if the .page files are already missing the info, it's on the g-ir-doc-tool side [11:19] http://people.canonical.com/~xnox/yelp/doc/Gtk.FontChooser.page looks really poor [11:20] pitti: right, so the Notify.gir works here. So not sure what's wrong with Gtk3. [11:20] * xnox really needs/wants Gtk, GLib, Gdk. [11:23] GdkX11 worked, Gdk/gtk/glib didn't =( [11:25] pitti: i'm on raring, are you? [11:25] xnox: yes, raring du jour [11:42] seb128, around? [11:48] seb128: ok, I demystified bug 1067876 [11:48] Launchpad bug 1067876 in gvfs (Ubuntu) "Missing "Safely Remove Drive" option from Quicklists. Only have "Eject"." [Low,In progress] https://launchpad.net/bugs/1067876 [11:48] seb128: it's again a "hardware lies, we have no information what's right" case, and really tricky to work around [11:48] seb128: but for now, showing "Eject" for USB sticks doesn't break the world IMHO [12:02] tkamppeter, hey, sorry I was away for a bit [12:02] pitti, oh, ok, thanks for figuring it out! [12:03] seb128, I have a problem with GNOME's printer setup tool and do not know how to debug it. [12:03] tkamppeter, what sort of problem? [12:03] seb128, I am on Raring and if I run system-config-printer from the command line (old UI) it detects all my printers, in the network and on USB. [12:05] seb128, if I use the new tool on my laptop (real-iron Raring), I see only network printers and no USB printers, looks like the new tool only checks via Bonjour but does not ask CUPS. [12:05] seb128, on a Raring VM no printers are found at all, probably also not asking CUPS and not finding anything via Bonjour as it has its own network inside my PC. [12:06] could be, did you look at the code? or maybe ask upstream (I think he's mkasik on the GNOME IRC) [12:06] need to be away a few minute, I will read the scrollback [12:06] brb [12:06] seb128, so the problem seems that the new tool is not asking CUPS (via s-c-p D-Bus service) to detect printers. === MacSlow is now known as MacSlow|lunch [12:11] Sweetshark: libcmis is uploaded. thanks for your work on that. i will review 4.0.1 [12:12] bdrung: please hold back uploading 4.0.1 a bit still [12:12] okay [12:14] bdrung: some test have been proven to be flaky in general (as they failed on the backports builders too). So Ill disable them, so that we dont end up with a 24 hour armhf build that then fails to complete. [12:14] bdrung: you can go ahead and review 4.0.1 already though as this will only be a minor change. [12:14] Sweetshark: "new upstream release candidate" -> "new upstream release" [12:26] tkamppeter, have you looked at the GNOME tool's code to see what it's doing exactly? [12:26] seb128, not yet. [12:27] tkamppeter, it seems like they use org.opensuse.CupsPkHelper.Mechanism DevicesGet method === Ursinha_ is now known as Ursinha [12:34] seb128, how do I fire up the printer tool from the command line, to see its warnings, and is there perhaps a debug mode for more verbose console logging? [12:37] tkamppeter, try "G_DEBUG_MESSAGES=all gnome-control-center printers" [12:37] but it doesn't seem like there are lot of debug code in the source [12:40] seb128, thanks for the hint, there is no output at all. [12:41] bdrung: fixed in http://anonscm.debian.org/gitweb/?p=pkg-openoffice/libreoffice.git;a=commit;h=36f27eb52d7c35b9a0ff59fb9298c11a3f64759a which also has the disabling of tests by ricotz. http://people.canonical.com/~bjoern/libreoffice4/ is updated too. [12:43] seb128, I reported https://bugzilla.gnome.org/show_bug.cgi?id=695286 now. [12:43] Gnome bug 695286 in Printers "Printer tool of GNOME Control Center does not get info about available printer devices from CUPS" [Normal,Unconfirmed] [12:43] tkamppeter, did you figure out what is wrong or just reported the issue? [12:45] bdrung: also, its official now: http://blog.documentfoundation.org/2013/03/06/the-document-foundation-announces-libreoffice-4-0-1/ [12:46] tkamppeter, you might want to join irc.gnome.org #control-center and ping mkazik to discuss the issue [12:50] is it just me or bzr is acting a little bit strange? [12:56] Sweetshark: for what reason does libreoffice-help-en-us conflict with all other help packages? [12:57] bdrung: bug 957589 [12:57] Launchpad bug 957589 in Ubuntu Translations "Localized LibreOffice Help files ignored when help for en-US is installed" [Medium,Triaged] https://launchpad.net/bugs/957589 [12:59] thanks [13:02] seb, thanks for the hint, I am connecting now. [13:07] re [13:08] ogra_, what's the recommended way to put raring on the nexus7 nowadays? did you say usb-creator works for it? === ara_ is now known as ara === MacSlow|lunch is now known as MacSlow [13:17] bdrung, hi, will you upload the new libreoffice within the next hours? [13:37] good morning, hax0rs [13:38] ricotz: i will review it and upload, but i can't promise it to be in the next few hours [13:38] attente, https://apps.ubuntu.com/cat/applications/precise/super-hexagon/ [13:38] mterry: nooooooo [13:38] hey desrt mterry [13:39] seb128: hey [13:39] seb128, desrt: Morning! [13:39] seb128: i took your advice [13:39] https://launchpad.net/~desrt/+archive/ppa [13:39] testing now [13:39] great [13:39] mterry: oh cool [13:40] seb128: is there anyone we can talk to about PPAs effectively being overnight delivery now? [13:40] you mean? [13:41] build time issues? [13:41] yes [13:41] i386 backlog is down to 1h40 [13:41] waiting 12 hours for a trivial build is not the best thing for productivity [13:41] I think that's going to solve itself in a bit [13:41] oh. good. [13:42] we accumulated delays for some reasons [13:42] but builders are back and catching up [13:43] hm. looks like i screwed something up. [13:44] oh. dailies. [13:45] bdrung, ok, thanks [13:51] So there was a session on rolling release yesterday. I looked through its notes, but didn't see anything interesting. Now an email to the ubuntu-desktop list says we decided to release 13.04 after all? What happened at that session? [13:52] Sweetshark: the changelog order is wrong. 1:4.0.0~beta2-0ubuntu2 should come before 1:4.0.0-1 [13:52] mterry, you can watch the video on youtube, nothing was decided [13:53] seb128, that would take time :) [13:53] mterry, basically there is a good part of the people who feel like we commited to have a 13.04 and it's late to bail out from that commitement [13:53] mterry, but there was no conclusion [13:53] mterry, it seems like middle ground was to do a release but only support it for security [13:53] e.g not do SRUs [13:54] suggest to people who want the fixes to just keep rolling on the rolling release from there [13:54] seb128, you will participate in the printing session in 5 minutes? [13:54] pitti, will you participate in the printing session in 5 minutes? [13:55] seb128, well, it is supposed to, otherwise use the manual way via fastboot [13:56] (usb-creator for flashing) [13:56] tkamppeter: yes, I'll listen in [13:56] ogra_, ok, let me try that ;-) [13:56] unzip the .gz first [13:57] mterry: you wrote parts of the datetime panel, right? [13:58] particularly the part for turning datetimes into a number to stick into a spinbox and using that number to render the date/time values [13:59] anyway.. i was testing my recent changes there [13:59] so i changed the time manually... that worked OK [13:59] Unity sucks [13:59] then i tried changing the date [14:00] that didn't go so well [14:00] GeorgeTorwell: i know, eh? seriously! [14:00] Yeah it sucks so hard bro. Cinnamon ftw [14:00] GeorgeTorwell: now that we've covered unity let's move on to why you don't like your science teacher [14:00] bdrung: no, changes need to be ordered in the way they appeared in the ubuntu releases. thus if 1:4.0.0~beta2-0ubuntu2 already merged changes from 1:4.0.0-1 in packaging the order has to be as is. [14:01] desrt: I'm not a neckbeard like you. And no amount of passive aggressive neckbeard rage is going to change the fact that unity sucks and it sucks hard. [14:02] GeorgeTorwell: i think i was suggesting that you were too young to grow a beard, actually [14:02] but the fact remains that you're the one here making useless remarks, obviously to very much the wrong channel [14:02] Sweetshark: 1:4.0.0~beta2-0ubuntu2 came directly after 1:4.0.0~beta2-0ubuntu1 and did not merge the _full_ changes from 1:4.0.0-1 [14:02] And now Ubuntu wants to come up with Mir, its own replacement for X instead of just using wayland? Well shit on a stick (ironically this is also desrt's nickname), how buggy is that going to be [14:03] GeorgeTorwell: in fairness, X kinda sucks [14:03] desrt, yeah [14:03] in fact, X sucking is a good part of the reason for compiz sucking which is, in turn, a good part of the reason for unity sucking [14:04] X does suck [14:04] mterry: can you reproduce the issue i describe? [14:04] but holy shit there's this thing called WAYLAND [14:04] Sweetshark: cherry-picking is different to merging [14:04] which Ubuntu was supposed to ship with in the future [14:04] GeorgeTorwell: can you please tell me what wayland is? [14:04] the replacement for X, and the future of linux [14:04] right [14:04] so if you could go ahead and arm yourself with some more facts before you start coming in here and babbling, that'd be great [14:05] wayland is just a protocol, you know? [14:05] desrt, did you edit the date text or use the up/down buttons? [14:05] no shit faggot, I said it was a replacement for x [14:05] mterry: up button, specifically [14:05] GeorgeTorwell: that language is utterly unacceptable. please leave. [14:06] lolno [14:06] I just popped in for a sec [14:06] desrt, works for me. You may have to wait a few seconds for the change to show up (not sure why on that off the top of my head) [14:06] hey jasoncwarner_! [14:06] hey jasoncwarner_ === m_conley_away is now known as m_conley [14:06] jasoncwarner_: hey. how's your back? [14:06] jasoncwarner_, tkamppeter: is the session actually happening? IRC is quiet and video doesn't work [14:07] mterry: hm. when i change it, i find myself in 1970 :) [14:07] even after ppa-purge and a reboot [14:07] bdrung: a linear changelog cant handle branching and merging properly, but if you insist on keeping that futile illusion ... [14:08] desrt, not the worst place to be, but not expected, no [14:08] :) [14:08] bdrung: dpkg-mergechangelog merges everything :X [14:08] lemme toss this guy under G_DBUS_DEBUG and see what's going on [14:12] Sweetshark: doko took 1:4.0.0~beta2-0ubuntu1 and did one change and uploaded it as 1:4.0.0~beta2-0ubuntu2. after the 1:4.0.0~beta2-0ubuntu2, you merged the changes from Debian [14:12] hm... no D-Bus involvement. quite odd! [14:12] maybe this chronology didn't end up in the same way in git [14:12] kinda takes all of the parts i've spent the last day or two breaking out of the picture [14:17] bdrung: http://anonscm.debian.org/gitweb/?p=pkg-openoffice/libreoffice.git;a=blobdiff;f=changelog;h=0ff1ae001f6e7c7d0e979f48e4dd0b7ef1ec6304;hp=0dd26df29d611f408a75c28f2d17742f5d49bb2c;hb=42dd5bc7d7fbd2b2ac410eb39fd114aab072ead0;hpb=574cc9f0fbaf8b1d7a4dc1bc287cda55403be432 [14:19] bdrung: yes, doko sponsored a changed libreoffice package to raring, without pushing his changes to DSCM or even notifying anyone about, leaving others to clean up the mess for his. :/ [14:19] s/his/him/ [14:22] Sweetshark: you can simplify "(LP: #785518) (LP: #949997)" to "(LP: #785518, #949997)" [14:22] Launchpad bug 785518 in libreoffice (Ubuntu) "[Upstream] soffice.bin crashed with SIGSEGV in SdrEndTextEdit()" [Medium,Fix committed] https://launchpad.net/bugs/785518 [14:22] Launchpad bug 949997 in libreoffice (Ubuntu) "[Upstream] Ctrl + z and Enter soffice.bin crashed with SIGSEGV in SfxLinkUndoAction::~SfxLinkUndoAction()" [Medium,Confirmed] https://launchpad.net/bugs/949997 [14:28] ls [14:28] ups [14:40] mterry: lol. [14:41] mterry: this is one of those "only in canada" bugs [14:42] mterry: in en_CA, %x (which is the format you use for printing the date) gives a 2-digit year [14:43] so when you parse it back again, we find ourselves in the year .... 13 [14:43] that results in GDateTime giving a rather large negative value for the unix timestamp [14:43] which the spinbox then truncates to zero [14:43] and.... ta da! 1970 [14:44] desrt, hrm [14:44] here's one thing i don't understand: you use %x for the date, which is locale-specific [14:44] so should be mm/dd/yyyy in US, right? [14:44] but you parse it back in again in the form "yyyy-mm-dd" always [14:45] desrt, let me look at code again [14:45] ya.... it fails to parse US-formatted times [14:45] and in canada it works but.... the result is not nice [14:46] desrt, it works here... [14:46] mterry: are you seeing a mm/dd/yyyy formatted date? [14:46] when i set LC_ALL=en_US that's what i get [14:47] and when i try to change it i get a g_warning to the effect of "WARNING **: Could not understand 03/07/2013" [14:47] desrt, yeah I do [14:47] and the date does not change [14:48] mterry: reading the code, there is really no way this can work.... [14:48] 28.6.15 mike@mt | scanned = sscanf (text, "%u-%u-%u", &year_in, &month_in, &day_in); [14:49] desrt, I am also confused on that line [14:50] so look here: http://bazaar.launchpad.net/~indicator-applet-developers/indicator-datetime/trunk.13.04/revision/120.1.1 [14:50] jjardon: you broke it :) [14:51] desrt, I get those print warnings. Then my system time changes... Not sure why yet [14:51] used to be hardcoded to %Y-%m-%d [14:51] so full 4-digit year, month, day [14:51] which would work just fine, always [14:51] desrt, ah yes. I hard coded it because I could not find a way to parse back a date in a generic locale format [14:52] * desrt would be annoyed to still be hunting down y2k bugs in 2013 if this was not actually a y0.1k bug [14:52] desrt, I blame Canada [14:52] mterry: i don't believe you that this works in the USA [14:52] it's not possible :) [14:53] mterry: to confirm: we are talking about manual text edit -- not using the buttons [14:53] desrt, this had been part of my subtle attempts to force ISO 8601 on the world [14:53] desrt, oh. I was using button [14:53] ah [14:53] that one works [14:53] unless you're me [14:53] desrt, I still got the warning, but then it worked [14:53] because the button causes the formatted string to update which parses successfully as the year 0013 [14:53] right [14:54] anyway.. there are two bugs here [14:54] we need to revert jjardon's commit [14:54] but then we probably also need to change the canadian preferred locale datetime to include a 4-digit year [14:54] desrt, good luck changing anything locale related [14:54] desrt, that is an uphill battle from what I've seen [14:55] sigh. [14:55] so revert jjardon's patch? [14:55] desrt, though I agree 4-digit years does seem better in this day and age [14:55] desrt, is there a bug yet? [14:55] nobody in canada writes a 2-digit year in ISO format [14:55] https://bugs.launchpad.net/indicator-datetime/+bug/729056 [14:55] Launchpad bug 729056 in indicator-datetime "Date setting in the preferences should be not be shown in ISO format" [Low,Fix released] [14:55] if you're a fan of reusing the bug that was used to introduce the regression [14:55] otherwise, no [15:00] seb128: so i think we're ready to go [15:00] except for this (unrelated) issue, everything works as expected [15:00] desrt, ok, great, "just" need systemd MIRed and systemd-shim uploaded, NEWed and MIRed [15:01] lovely :) [15:01] * larsu just realized that MIR now means two things... [15:01] larsu: we have Mir and MIR [15:02] fair enought [15:02] *enough [15:02] [MIR] mir will be my favourite bug [15:05] tedg: did you follow this conversation? looks like you were the one to approve the merge? [15:06] tedg: probably the easiest fix right now is a revert, unless you want to do a lot of work [15:06] desrt, commented on that bug and filed a new one too [15:06] k [15:06] mterry: you gonna take care of it, then? [15:06] desrt: A bug come from the past to bite me! ;) [15:07] Laney, now now now, it would be "[MIR] Mir". All lower-case "mir" is still an available acronym! [15:07] I was thinking of the package name :P [15:07] mterry: you could get fancy... [15:07] desrt, I want to quickly poke mpt about it. mpt: feel like looking at bug 1149696 ? [15:07] Launchpad bug 1149696 in indicator-datetime (Ubuntu) "Manually editing date field results in bogus date" [Undecided,New] https://launchpad.net/bugs/1149696 [15:08] mterry: you could try for format january 2nd 2003 [15:08] Laney, ah, yeah you're right [15:08] and see if it comes out like 01-02-03 or 01-02-2003 or 02/01/2003 or whatever [15:08] then decide what the format is based on that [15:08] desrt, but there are things like Canadians [15:08] i think it's also possible to query nl_langinfo for the format string corresponding to %x [15:08] so you could use that to scan with [15:08] desrt, oh really? I don't recall that being the case [15:09] glib does this, i think [15:09] that's how g_date_time_format() works [15:09] desrt, I *believe* I looked into this in the past and couldn't find it. But maybe I overlooked it [15:09] PREFERRED_DATE_FMT [15:09] Oh, in that case... [15:09] bdrung: http://people.canonical.com/~bjoern/libreoffice4/ updated [15:09] #define PREFERRED_DATE_FMT nl_langinfo (D_FMT) [15:09] ding ding [15:10] it's still a lot of work to write a parser based on this [15:10] and note that you may not even be dealing with ascii digits [15:10] i think in a farsi locale you get the arabic arabic digits [15:11] desrt, wait, farsi doesn't use "western" arabic digits? I always assumed we got those from them [15:11] mterry: we got the system from them, but not the glpyhs [15:11] desrt, cool [15:11] ٠‎ - ١‎ - ٢‎ - ٣‎ - ٤‎ - ٥‎ - ٦‎ - ٧‎ - ٨‎ - ٩ [15:12] ironically, the arabs call these "hindi numbers" [15:12] because they came from india originally, no? [15:13] yes [15:13] but the indians write them like ०.१.२.३.४.५.६.७.८.९ [15:13] desrt, hmm. I don't want to write a generic date parser. I vote ISO 8601 [15:13] mterry: ya. me too. glad you agree. :) [15:13] * mterry wonders if that was the real reason I bailed way back when [15:13] mterry: this is a job that could easily consume a week of your time dealing with edgecases [15:13] i don't think having locale-specific date formats is worth a week of your time [15:14] desrt, also, are you sure there aren't differences between %x and D_FMT...? [15:14] mterry: well [15:14] Sweetshark: some changelog comments: there is a typo in "diable help building on armel/armhf" -> it should be "disable" [15:14] mterry: you use g_date_time_format() [15:14] which implements %x using D_FMT [15:14] so.... no. [15:14] fair [15:14] no difference :) [15:14] Sweetshark: you can drop the "tweak Maintainer field" changelog entry [15:15] desrt, mterry: what does ISO 8601 specify? [15:15] seb128: date and time formats [15:16] 2013-03-06T14:16:00 [15:16] seb128, specifically, this is YYYY-MM-DD [15:16] among many many other things [15:16] that seems wrong [15:16] yyyy-mm-dd is the international standard [15:16] we would here DD-MM-YYYY here for example [15:16] seb128, it's the only thing that's right in a world of evil [15:16] because, you know, canada uses it [15:16] seb128: no. you would not. [15:16] seb128: you would do dd/mm/yyyy [15:16] with / [15:16] works for me [15:17] - is reserved for people with sane systems [15:17] as long as days are first [15:17] ie: canadians [15:17] seb128: YYYY-MM-DD is the only date format that cannot be mis-interpreted [15:17] if days come first, so should the time: 18:37 05/02/2000 [15:17] larsu: you mean 37:18, right? [15:17] mterry, are you going to tell me that GTK still doesn't have a datepicker widgets? [15:18] desrt: :D [15:18] seb128: doing it the other way around (year first) makes it sortable, though [15:18] bdrung, there is no mis-interpretation in a specific locale [15:18] seb128: as much as i love little endian, you europeans have the date thing totally wrong :p [15:18] * larsu agrees [15:18] larsu, I'm not arguing in favor, I'm just saying that users expect it this wat [15:18] way [15:18] I'm surprised no-one has cited XKCD yet [15:18] the americans do too, of course [15:19] seb128: okay. the mis-interpretation can only happen if you don't know the locale [15:19] mpt, I went to that comic to remember the ISO number :) [15:19] larsu, you can claim that the europeans users are stupid and don't care but they are still our users [15:19] seb128: if users see 2013- out in front they know that they should alter their expectation [15:19] Sweetshark: why where a build-conflict on gcc/g++ (>= 4:4.7~) [!i386 !amd64 !kfreebsd-i386 !kfreebsd-amd64] added? [15:19] seb128: and this is not some theoretical "this way or that, all the same" discussion [15:19] desrt, or they complain that the format is weird [15:19] this is a "mterry will have to spend a week fixing this for you to have it your way" discussion [15:19] seb128: yeah, sadly. I didn't say users were stupid, only that I prefer it the other way around [15:19] bdrung: finished? if not, can you collect the needed changes when you have all of them so that I dont cycle around endlessly doing trivial changes. [15:20] desrt, yeah but then we fix it real nice for everyone forever :) [15:20] Sweetshark: nope. not finished with my review. [15:20] larsu, what we prefer doesn't really matter though... [15:20] mterry: if you write this code, i want it in GDateTime :) [15:20] desrt, mterry, larsu: doesn't seem like spending lot of time on though [15:20] bdrung: because we had them before, I dont recheck the debian delta on each upload. usually I purge the delta as far as possible for the next libreoffice major [15:21] seb128: I know, but I want to state my opinion regardless :) [15:21] seb128: the problem is that we have a library function to format dates according to the current locale, but none to parse them back [15:21] mpt, no, I don't believe they do have a date picker. Let me double check GTK hasn't added one since I last looked [15:21] Sweetshark: have you done a testbuild on non-i386/amd64? i am not sure what this new build conflict means for non-i386/amd64 [15:22] mpt, just their calendar widget [15:22] ffs [15:22] mterry: you want a calendar attached to a label or something? [15:22] like, airline booking site style? [15:22] i bet writing one of those is easier than writing a generic date parser :) [15:22] desrt, mpt: true, we could adjust the design so it isn't directly editable [15:22] Fifteen years and still no datepicker [15:23] also: wtf is the deal with the spin buttons [15:23] i press up and it changes *minutes*? [15:23] desrt, hold it down and it goes faster [15:23] heh [15:23] I think... [15:23] thank god for ntp [15:23] The spin buttons should increment/decrement the selected bit. Currently they increment/decrement ... something. [15:23] Minutes and days. [15:24] Sweetshark: the build conflict on gcc/g++ (>= 4:4.7~) is new [15:24] mpt, hah, I missed that part of the spec (to increment the selected bit) [15:24] selected? as in 'the one the cursor is inside of'? [15:25] mterry, I didn't specify it, because it's a standard part of a datepicker :-] [15:25] mpt: isn't a datepicker that lets you choose a date from a calendar much more helpful than a spin box? [15:26] larsu: that's kinda what we're bitching about just now [15:26] larsu, it's much slower, especially if the date is way off, e.g. 1970. [15:26] But other than that, yes. :-) [15:27] but that's an edge case. Who ever sets their time 40 years into the past? [15:27] larsu: happened to me today, quite a few times :) [15:27] larsu, it's one of the most likely wrong years, being Unix year zero [15:27] I knew someone would say that :D [15:27] bdrung: checking ... somebody did suggest whitespace changes to debian, making merges a real pain ... [15:27] mpt: ah, good point [15:27] ;) [15:28] in that case, a text field is probably best... [15:28] * larsu prefers NTP anyway [15:28] ya. thank god for ntp. [15:28] mterry, so, how do you feel about implementing the datepicker if I specify it in detail? [15:28] Sweetshark: these changes made reading debdiff more easily :) [15:28] * desrt is amazed that in all of the option-removing zeal, GNOME hasn't stripped away the option to disable network time [15:28] because, seriously... who actually wants their time to be wrong? [15:28] people who aren't connected to the internet [15:29] oh. Nevermind. [15:29] actually... i know a lot of people who set their alarm clocks fast [15:29] mterry, basically a text field, where you can navigate between its segments with Left+Right keys, increment/decrement with Up/Down, and type a complete replacement date without having to tab or type separators. [15:29] mpt, I'm fine with it. But the priority is a different question. It's not mobile related, so... :) [15:29] true [15:29] mterry: just commit the revert [15:29] mpt, also it'd be difficult to get right due to the variety of parsing back locale-specific formats, so it would take some time [15:30] it seems non trivial work for little benefit in the big picture [15:30] e.g very low priority [15:30] let's go with the easier fix [15:30] mpt, how much do you hate going back to ISO 8601? [15:30] e.g revert the commit? [15:30] seb128: yes [15:30] we can leave the old bug open in case some day someone wants to make the effort [15:31] This time I'm going to leave a comment, pointing at these bugs [15:31] mterry: good idea :) [15:31] bdrung: sure, I hear you want to do the merges there now. [15:31] ;) [15:31] mterry, I love ISO 8601 with a passion. I use ISO 8601 on all my design sketches. But I usually have to explain it if I dare use it with a non-geek. [15:32] mpt: you should move to canada [15:32] we live and breathe iso 8601 here [15:32] Sweetshark: the whitespace change should have only created a one time merge complication [15:33] desrt, or Lithuania. The locale settings in Lithuania are awesome. [15:34] * desrt would have assumed they used the same as the rest of europe [15:34] desrt, your locale says otherwise, two-digit-year man [15:34] mterry: i told you this is another bug [15:35] bdrung: yep, that is a merge error, moving the Build-Conflicts: from debian to us. [15:35] desrt, :) [15:35] desrt, in my opinion, it's a bug with all the locales that they aren't just iso8601 [15:35] But that's the programmer in me talking [15:36] mterry: this is the kind of bug that would be closed 'Opinion' :) [15:36] mterry: let's next move on to how daylight savings time is only used by morons and how we should all be on UTC as well :) [15:36] mterry, perhaps a cheap way of improving the situation would be to add a calendar button next to the date field. [15:37] bdrung: since gcc-4.6 is in universe that would likely result in a component mismatch. Ill remove that conflict again. [15:37] https://code.launchpad.net/~mterry/indicator-datetime/iso8601/+merge/151989 for review [15:38] desrt, pfft, UTC is so Earth-centric. I bet there are all sorts of cool systems that work in space and at different velocities [15:39] Ubuntu could be a real leader here! :) [15:39] mterry, I doubt it, because relativity [15:39] mterry: i cite our motto as an appropriate scope [15:39] linux for HUMAN BEINGS [15:41] mpt, stupid relativity making it hard to fix bugs [15:42] mterry: approving... [15:43] cyphermox, ooh thnks [15:43] desrt: if there's a lot of prospective martian users, we might want to include them too [15:43] but then we'll have to rename the distro [15:43] or at least change the tag line [15:44] Yeah, seems specist right now (specie-ist?) [15:44] haha [15:44] bdrung: 99c0770a6c5033f39649716f353cdefdf55b9379 removed conflicts against 4.7 so we had quantal on 4.7, so it should be good. [15:44] my cat even tends to favor walking on my key [15:44] GNOME had a participant in the womens outreach programme who was offended by this [15:44] *keyboard rather than others [15:44] desrt: by what? [15:44] 'linux for human beings' [15:45] she was also offended by the term 'human rights' [15:45] ah [15:45] Sweetshark: so this change got lost in the merge [15:46] would have loved to hear the reasons why, but this is *so* off topic for here [15:47] bdrung: yep, I already removed the conflicts again, but wont push until you completed your review. [15:49] Sweetshark: you seem to have lost the ttf-sil-gentium-basic -> fonts-sil-gentium-basic change in the merge [15:55] bdrung: fixed [16:02] Sweetshark: wrap-and-sort change: paste.debian.net/240140/ [16:02] http://paste.debian.net/240140/ [16:03] you can recreate it (on raring) with: wrap-and-sort -d . -v && for i in control*.in; do echo >> $i; done [16:10] bdrung: hmm, I did a wrap-and-sort bluntly on ./debian. It also touched ./debian/copyright ... [16:10] * Sweetshark checks if that is sane. [16:10] Sweetshark: just revert the copyright change [16:10] it trims of trailing spaces [16:11] Sweetshark: the copyright file should be fixed in the experimental branch and then merged [16:12] it follows an old format specification and it doesn't follow the guidelines [16:13] it should use Format: http://www.debian.org/doc/packaging-manuals/copyright-format/1.0/ [16:17] Sweetshark: control.in: libreofficeVER-writer has a trailing comma in Recommends [16:17] Sweetshark: "integration and a GConf backend" -> two spaces instead of one after "integration" [16:18] Sweetshark: i finished the review of the debian/ diff. now i will check lintian and do an install+using test [16:21] bdrung: fixed (trailing , was already fixed by wrap-and-sort) [16:27] me wonders what happened to the builders -- 9 hours delay before amd64 starts building ... [16:44] amd64 21 277 jobs (7 hours 20 minutes) === mitya57_ is now known as mitya57 === francisco is now known as Guest71447 === mhr3__ is now known as mhr3 [19:12] desrt, http://youtu.be/VTlPLZfJDVc [19:12] desrt, migration to logind session if you want to listen in [19:12] desrt, #ubuntu-uds-client-1 on IRC === rickspencer3_ is now known as rickspencer3 [19:48] desrt, you missed it [19:49] Sweetshark, yeah amd64 has been getting swamped now and then for me [19:57] * didrocks waves good evening [21:59] thomi: for autopilot, i can't type into gedit when launching it using launch_test_application, while doing start_app_window works fine [22:01] attente: Hi - I wonder if alesage can help you with this - I need to catch up on sleep :-/ [22:01] * thomi quits IRC [22:02] thomi: no worries, thanks [22:02] you're stuck with me attente :) [22:02] aw drat :) [22:05] alesage: maybe i should first ask what the difference between the two are [22:06] attente I believe one uses the Unity native bindings and the other the autopilot-gtk bindings [22:06] and I believe that typing is possible via autopilot-gtk, just trying to demonstrate to myself [22:08] attente, which tests are you working on? [22:10] a test i haven't pushed yet [22:17] I'm attempting to install egg-list-box using jhbuild and am making it as far as the build phase before I get the error "/bin/bash: --pkg: command not found". The full output from that phase can be found at http://paste.ubuntu.com/5591640/. I can't find any package called 'pkg' (unless someone's misspelled dpkg). Can anyone help me out with this? === m_conley is now known as m_conley_away [23:21] is there a desktop beer hangout? =) [23:21] notgary: you should ask in a GNOME channel about that [23:32] jbicha, thanks. I've managed to figure it out anyway. I was missing the Vala build tools. I'll pester the gnome people next time about it :P