[00:00] robert_ancell: Did metacity get sponsored? If not, I'll take care of it this morning. [00:01] TheMuso, didrocks did it, thanks [00:01] np [01:46] TheMuso, can you sponsor compiz-fusion-plugins-main from lp:~compiz/compcomm-plugins-main/ubuntu [01:46] sure [01:47] do you have any idea why debian/install stopped working (and now requires debian/tmp in there)? [01:56] not without looking deeper into the package history [01:57] what is the "correct" way - should all install files have debian/tmp ? [01:58] As far as I understand things, if there are multiple packages being created from one source package, all files go into debian/tmp and are copied to their apropriate packages from there. [01:59] that sort of thing can depend on debhelper compat level [01:59] dh_install in compat mode 7 will automatically look in debian/tmp. [01:59] ^^ what he said [02:01] ah, thanks. It must have transitioned to 7 [02:10] robert_ancell: uploaded [02:11] TheMuso, thansk [02:11] * TheMuso must admit he is a little distracted by today's Australian political machinations. [02:11] And, disturbed. [02:11] macinations even [03:35] anyone around who can look at a gobjection-inspection patch? [03:58] NCommander: would you happen to know what source the gnome shutdown dialog is in? [03:58] micahg: gnome-session I think [03:58] been awhile since I looked [03:58] NCommander: k, thanks [04:14] Gah! Damn you launchpad. Don't oops on bug submission. [04:14] Thank you Chromium, and your delicious form-saving back button. [04:29] <3 [04:29] RAOF: do you need any help upstreaming those f-spot patches? [04:29] or is that pretty much good [06:28] Plenty of bugs for indicator-network! [06:34] RAOF: hah [06:35] But my UltraBase™ has arrived, and displayport is pleasantly painless. [06:36] ooh [06:36] hey [06:36] I'm not entirely sure how people actually _use_ VGA outputs. Or maybe my cable is made of rusty iron. [06:36] can you do dual hdmi or whatever it is the cinema displays need, on an x201s with UltraBase ? [06:36] RAOF: rusty iron. [06:36] RAOF: I will say that I hate vga + LCD. But my 21" philips professional CRT kicks arse. [06:37] I haven't had a CRT for over a decade. [06:37] Actually, that's not quite true. I picked one up from the side of the road to test VGA output at one point :) [06:38] lifeless: Displayport can certainly push out the needed bandwith for your choice of insanely sized monitor. I'm fairly sure the software will also be up to it. === almaisan-away is now known as al-maisan === al-maisan is now known as almaisan-away [07:33] RAOF: Does it also put audio out via displayport? [07:36] I don't know. The displayport protocol certainly allows for that, but I don't have anything that accepts DP audio. [07:36] The DP protocol is actually rather freaky. You can do USB over DP if you feel like having a USB hub in your monitor. [08:19] hey guys [08:20] micahg proposed to get gjs rebuilt (bug 597944) for gnome-shell [08:20] Launchpad bug 597944 in gjs (Ubuntu) "libgjs needs a rebuild for xulrunner-1.9.2.4 (affects: 1) (heat: 8)" [Undecided,New] https://launchpad.net/bugs/597944 [08:20] I wanted to sponsor it, but for me it seems to FTBFS [08:20] http://paste.ubuntu.com/454309/ [08:26] hey seb128 [08:26] salut mon ami - comment ça va? [08:26] hello dholbach [08:26] dholbach, ca va bien merci, et toi ? [08:27] ca va bien aussi :) [08:27] dholbach, qu'est-ce qui t'amène chez nous ? un bug de nouveau ? ;-) [08:27] micahg proposed to get gjs rebuilt (bug 597944) for gnome-shell [08:27] Launchpad bug 597944 in gjs (Ubuntu) "libgjs needs a rebuild for xulrunner-1.9.2.4 (affects: 1) (heat: 8)" [Undecided,New] https://launchpad.net/bugs/597944 [08:27] I wanted to sponsor it, but for me it seems to FTBFS [08:27] http://paste.ubuntu.com/454309/ [08:27] anything I should be doing about it? (seems he went to bed now) [08:27] I dunno if it was important or urgent or anything [08:29] dholbach, I was just reading my email, got one from him about that [08:29] ah ok, well, I guess I just update his bug then [08:29] dholbach, no, don't bother I meant to have chrisccoulson to give a try to it today or maybe do the merge from debian [08:29] they have a new version of gjs there [08:30] ah, gotcha [08:30] dholbach, thanks for asking about it though ;-) [08:30] yeah, I didn't know if it was important in any way [08:30] it's not [08:30] it's required to have gnome-shell new version to build [08:31] ah ok [08:31] but it's not something we rely on, it's in universe [08:31] so if it takes a few extra days that's not an issue [08:31] alrightie [08:31] good morning [08:32] seb128: je suis près de toi maintenant :) [08:32] (seems like I overslept :)) [08:32] lut didrocks, nice! [08:32] salut seb128 :) [08:32] dholbach, oh ? en vacances ? [08:32] seb128, does that look sane to you ? http://launchpadlibrarian.net/50829991/gobject-introspection.debdiff [08:32] does anyone know how to make dpkg triggers work? I have a working dh_gsettings, but I think I should reimplement as a trigger [08:33] (i know its just a workaround) [08:33] seb128: non, pas des vacances, mais je suis à Trèves :) [08:33] seb128: visiter mes parents [08:33] robert_ancell, create debian/.triggers [08:33] robert_ancell: I've had a look at them before, although I'm a bit hazy. [08:33] ogra, I'm not sure why that's required on arm and not other archs [08:34] ogra, could you open an upstream bug about it maybe to get their comments? [08:34] robert_ancell, you can take gconf or gnome-menus as an example [08:34] ogra, and where do I implement the trigger script? [08:34] robert_ancell, you need a .trigger and a trigger case in the postinst [08:34] robert_ancell, the postinst snippet has the code to run [08:34] seb128, will do, there is bug 597947 as downstream one already though [08:34] Launchpad bug 597947 in gobject-introspection (Ubuntu) "ARM FTBFS fix (affects: 1) (heat: 8)" [High,Confirmed] https://launchpad.net/bugs/597947 [08:35] seb128, so the code is shared between all packages? [08:35] robert_ancell, editor /var/lib/dpkg/info/shared-mime-info.postinst [08:36] robert_ancell, see the triggered case [08:36] robert_ancell, see /var/lib/dpkg/info/shared-mime-info.triggers to register what trigger the triggers [08:37] robert_ancell: yesterday, I had a look at metacity patches and saw you added "Description ? Authors?" to most of them. I've added those to 03_strict_focus.patch: if you google for META_FOCUS_MODE_STRICT, you'll find an IRC conversation between seb128 and lamont from Januay 2005 :) [08:37] robert_ancell, so you basically want a .triggers with your directory listed [08:37] robert_ancell, and the posting to run the update command in the triggered case [08:37] posting -> postinst [08:38] oh, so I make these changes in libglib2.0-bin, and then once you have the new version installed any program that touches files in /usr/share/glib2.0/schemas will trigger the script in libglib2.0-bin? [08:38] didrocks, did you see my comment about the "Desk" naming? can you drop those in the next upload [08:38] robert_ancell, right [08:38] much easier than modifying debhelper/cdbs :) [08:38] ;-) [08:38] seb128, would you accept the patch as a temp. workaround at least ? [08:39] robert_ancell, could you have a quick review of http://launchpadlibrarian.net/50829991/gobject-introspection.debdiff [08:39] robert_ancell: hi, I just look over the compiz merge diff. I'm sure that was a lot of work, good job! some nitpicking. i would really prefer to keep the old ubuntu changelogs, that should be trivial with dpkg-mergechangelogs. it also brings back compiz-gtk and remove compiz-dbg. I don't mind dropping compiz-dbg, but I don't think we need compiz-gtk (in addition to compiz-gnome). I have no strong opinion here though [08:39] seb128: not on #ubuntu-desktop? [08:39] didrocks, no [08:40] still, didn't finish to backlog everything :) [08:40] didrocks, ok, so "Desk" is a mistake, we dropped those changes from libwnck in lucid [08:40] mvo, I didn't realise at the time that we'd removed compiz-gtk. Why do you want the old changelogs? I think they just make it more confusing to work out what has changed [08:40] didrocks, your changelog explains why some people still get naming issues [08:40] didrocks, I didn't notice the wms had changes for that [08:41] seb128: ok, I will revert in mutter and in metacity so [08:41] didrocks, don't do an upload for that but next time you touch those... [08:41] robert_ancell: I think in the compiz case the old changelogs have a lot of useful information about the patches we added, who did it and when [08:41] didrocks, thanks [08:41] mvo, launchpad has the upload history as well [08:41] seb128: I have to do an upload with the papercut about Alt + tab and copy/paste [08:41] I will not argue about those [08:41] but I don't like keeping tons of useless changelog entries [08:41] robert_ancell: I can see that its not that useful for a package that is less heavily patched (e.g. plugins-main). but for compiz I think there is some value [08:41] it makes merges harder and takes CD space [08:42] seb128: I disagree that they are useless [08:42] seb128: and dpkg-mergechangelogs should make the merge trivial [08:42] ok, fair enough [08:42] as said I will not argue I know other people have other opinions [08:42] mvo, I moved all the patch info to patch headers [08:42] we never kept those for desktop packages [08:42] Good morning [08:42] hey pitti [08:42] hey pitti === ogra_ is now known as ogra [08:44] robert_ancell: yeah, that is a good thing! it appears that they also all contain the version the patch was added so I'm happy [08:44] grmbl [08:44] silly reconnects [08:44] ogra, I've no strong opinion about this change [08:44] patch-header info makes me happy to, so I'm happy now and don't want changelogs anymore [08:44] thanks :) [08:44] I don't know enough about gobject introspection to comment [08:44] mvo, ;-) [08:45] see, everybody always agrees with robert_ancell in the end [08:45] :) [08:45] I don't know how we does that [08:45] but I need to learn from him ;-) [08:45] heh :) [08:45] we -> he [08:45] robert_ancell, did you comment on the patch already (i had a reconnect) [08:45] robert_ancell, could you give a quick ack, nack, don't know on http://launchpadlibrarian.net/50829991/gobject-introspection.debdiff ? [08:46] ogra, no, but I just read that pygi was merged into python-gobject [08:46] robert_ancell: so about compiz-gtk - I don't mind either way, I doubt its a very useful pacakge, but you are the matinainer ;) [08:46] seb128, don't know [08:46] robert_ancell, thanks [08:47] ogra, ok, feel free to upload but open a bug upstream with the build error and your workaround [08:47] seb128, thanks a lot :) [08:47] np, thank you for working on it [08:47] mvo, my motivation is mostly just to have the same package in Debian and Ubuntu so we can make progress twice as fast [08:47] * ogra hugs seb128 ... also because he thinks french world cup viewers always need comforting ;) [08:48] ogra, oh, the patch seb128 just showed me? I hadn't seen it before then [08:48] ogra, hehe [08:48] robert_ancell: fair enough, it looks good I will sponsor if you want [08:49] robert_ancell, is pygi in pygobject a win, out of that making the pygi packaging work efforts somewhat wasted? [08:50] seb128, well, it meant we had a little bit of a head start in understanding pygi so that's not a bad thing [08:51] mvo, I agree we should remove the -gtk package, and then pester Debian to do the same if it's the right move [08:53] robert_ancell, btw gdm upstream seems very active atm [08:53] robert_ancell, we could maybe try to get some of our changes reviewed [08:54] robert_ancell, I noticed they fixed some of the issues for which we have distro changes, next update will be interesting [08:54] seb128, they just closed the "no gdmsetup" bug [08:54] WONTFIX [08:54] robert_ancell, right, they disagree it's the way to go, they want those options integrated in other configurations dialogs [08:55] dosen't stop you having a dialog in the mean time... [08:56] robert_ancell, right, I stopped trying to argue with them about this though, they don't see the point for fedora and will not consider it [08:58] seb128, was there any particular feature from newer gdm's that we want, maybe a patch candidate? [08:58] no [08:59] I've been backport quite some fixes to lucid [08:59] they are doing lot of refactoring in git atm apparently on widgets loading, etc to speed things up it seems [09:03] didrocks, btw vala fails to build on debian the same way [09:03] didrocks, you might want to tell slomo about it [09:03] seb128: yeah, I saw that [09:03] seb128: -2 as well? [09:03] yes [09:03] seb128: where do you see that? [09:03] slomo, hi [09:03] slomo, your vala updates don't build in i386 [09:03] (I saw for -1 as with the new upload, but not the build result) [09:04] didrocks, https://buildd.debian.org/fetch.cgi?&pkg=vala&ver=0.9.2-2&arch=i386&stamp=1277365652&file=log [09:04] seb128, hey, do you know how gio-querymodules works? I notice it's only run when new packages are installed (postinst). Shouldn't it also be run postrm? [09:04] seb128: thanks [09:04] robert_ancell, it's a trigger, it should run whenever the dir change no? [09:05] robert_ancell, watch the .cache in the /usr/lib/gio dir [09:05] seb128, so the postinst script always gets run when a trigger occurs? [09:05] robert_ancell, it should yes [09:05] ok [09:05] I've added strace to vala, tried to have a look but nothing relevant to me [09:07] brb restart [09:13] later all [09:15] seb128, didrocks: thanks... any idea what could be wrong? :) looks like a gio problem [09:16] slomo, not really no... [09:17] I tried to strace the check yesterday, nothing useful (in addition, I can't reproduce it in my pbuilder :/) === oubiwann is now known as oubiwann-away === oubiwann-away is now known as oubiwann [09:24] didrocks: same here :) maybe caused by a read-only $HOME? [09:24] slomo: hum, can be, good idea [09:28] didrocks: g_bus_get_sync() for the session bus is the call that fails [09:55] that nautilius bug has become too difficult to tolerate :/ [09:56] what? which one? [09:56] the one you just commented - the amount of begging, bitching, and complaining for something we didn't even do :/ [09:57] oh, right, just ignore it [09:57] brb [10:09] morning === tkamppeter_ is now known as tkamppeter [10:34] hey guys [10:34] I was wondering if you need more contributors for the Desktop team [10:39] dholbach: isn't that like asking "do you want more money"? i.e. few people would say no ;) ? [10:39] depends, is it counterfeit ? [10:40] :) [10:40] I was just wondering what kind of stuff desktop contributors should be doing [10:41] dholbach, triaging bugs, doing merges on debian, reviewing bugs with patches and sending those upstream or to debian, trying to lower our delta [10:41] seb128: hi , upstream just pushed a slightly different fix for Bug #270206 , just a heads up [10:41] Launchpad bug 270206 in rhythmbox (Ubuntu) (and 2 other projects) "RB should never start minimised to tray (affects: 32) (dups: 5) (heat: 156)" [Low,Fix released] https://launchpad.net/bugs/270206 [10:41] dholbach, triaging bugs, doing merges on debian, reviewing bugs with patches and sending those upstream or to debian, trying to lower our delta, doing updates, etc [10:41] vish, ok, is there an issue with the current one? [10:41] seb128: do you think somebody of the team could talk a bit about that at UDW? it'd be a good opportunity and we still have a few slots open: https://wiki.ubuntu.com/UbuntuDeveloperWeek/Prep [10:42] seb128: nope dont see any problems , just mentioning in case you didnt want any delta :) [10:43] vish, I do regular snapshots for rb so no worry [10:43] :) [10:44] dholbach, I knew you would try to trick me in! ;-) [10:44] dholbach, shouldn't the UDW topics be rather technicals? [10:46] seb128: merging stuff, updating packages, etc sounds technical enough to my ears :) [10:46] talking a bit about the process, maybe do a quick example or two [10:46] done :) [10:47] dholbach, do I need to have the topic decided now? [10:47] dholbach, I'm fine doing an UDW session but I want to think a bit about the topic [10:47] seb128: which slot on https://wiki.ubuntu.com/UbuntuDeveloperWeek/Prep would suit you? [10:47] dholbach, it will be something desktopish ;-) [10:48] dholbach, any of the slots on the 12th or 13th or the 8pm on the 15th [10:48] dholbach, let's say the monday 7utc one [10:49] seb128: just seb128 or seb128 and his desktop mafia? [10:49] just put my name for now [10:49] I will sort details in the next days and update people and title ;-) [10:50] super [10:50] seb128: https://wiki.ubuntu.com/UbuntuDeveloperWeek/Prep - done [10:50] dholbach, thanks [10:51] dholbach, it has better bringing me some contributors or I will ask for my money back! ;-) [10:51] seb128: sure :) [10:51] seb128: so… you're looking forward to the Germany game on Sunday already? ;-) [10:53] dholbach, seeing Germany getting kicked out by our british friends, sure ;-) [10:54] seb128: MAN! YOU'RE GERMAN! [10:54] * dholbach better leaves the channel quickly :) [10:54] dholbach, lol [10:54] * seb128 hugs dholbach [10:54] * dholbach hugs seb128 back [11:05] seb128: do you know how to set gconf key values in deb packages? [11:05] ah, I saw the game last night in a "Biergarten", very amusing [11:05] as in default values [11:05] tseliot: have a look at the ubuntu-artwork package [11:06] dholbach: will do, thanks [11:06] tseliot: that's done with the gconf schema files in general (/usr/share/gconf/schemas) [11:06] tseliot: or a debian specific "overrides" set in /usr/share/gconf/defaults/ [11:07] the latter is dh_gconf and generally easier to use in packaging [11:07] tseliot, you can add a .gconf-defaults in the debian directory with lines "key value" [11:07] binary.gconf-defaults [11:07] does anyone have a printer and a minute to grab a log for me? [11:07] seb128, pitti: ah, nice, thanks [11:07] pitti, on lucid or maverick? [11:07] seb128: doesn't matter [11:07] pitti, I can on lucid [11:08] I need an "udevadm monitor -e --udev" output when you plug it in [11:08] pitti, ok, one minute [11:08] I don't have a printer these days, sorry (my wife has it in Munich) [11:08] * pitti hugs seb128, merci [11:08] seb128: what I'm trying to do is to stop system-config-printer.py being started on boot [11:09] and instead launch it from update-notifier once either cups sends out a job notification or we get a printer hotplug event [11:09] -> avoids python in the ureadahead pack and boot sequence [11:12] mvo: hm, it seems update-notifier 0.99.3 is in lucid, but missing from lp:~ubuntu-core-dev/update-notifier/ubuntu [11:12] mvo: it does have versions up to 0.99.2 and then the debian/maverick ubuntu bits [11:14] pitti, http://paste.ubuntu.com/454372/ [11:15] mvo: 0.99.3 is in the /lucid branch, but shouldn't the fix also be in trunk? [11:15] pitti, do we need python there, I though we started those autostart with a delay [11:15] seb128: merci beaucoup! [11:15] pitti, de rien [11:16] seb128: we do start it with a delay [11:16] seb128: but all the python bits are still in the ureadahead pack [11:16] pitti, so do we need to profile it? [11:16] I know we are craving for sub-seconds here :) [11:16] pitti, well, couldn't we just drop those and pay a bit extra load after the delay? [11:16] but even boot speed aside, it's a constantly running python process [11:16] right [11:16] and we don't really need it until we actually print stuff [11:16] pitti: iirc the changes are not releavent for lucid -> maverick, its just a fix for a upgrade issue [11:17] mvo: I see, thanks [11:17] seb128: a cheesy hack would be to delay by 60 seconds, then it would be off the ureadahead collector radar :) [11:17] but still, we do not really need to have that running all the time [11:17] how many people print on a netbook every day [11:17] and for those it won't be worse [11:18] pitti, right, I'm not discussing the fix, but I was suggesting as a workaround if you care about login you could delay it a bit and drop it from the profiling, it will be slow to start but that will not count in the session [11:18] seb128: right; that's on the radar (bump the delay from 30 to 60 s) [11:18] pitti, yeah, the suggestion was just for a quick hack to get things working today [11:18] if you can get the nice way done as well that's better indeed ;-) [11:19] -as well [11:19] * pitti feels sorry for mvo to abuse update-notifier even more [11:19] pitti: what do you do this time? [11:19] you know, if only we had a system which could launch programs on certain events [11:19] we could call it "uplaunch" or "upstart" or so [11:19] pitti: and its all going into upstart, maybe even this cycle ;) [11:19] pitti: it will be run for the user as well [11:19] mvo: right, that's what I'm craving for [11:20] pitti: yeah - what will you do with u-n this time? [11:20] mvo: for now, u-n is just a convenient place to run system-config-printer.py when a printer gets connected or a print job is sent [11:28] mvo: would you mind if I rename the "firmware" bit to "uevent" in trunk and reorder things a bit so that it's easier to add new uevent types? [11:28] then I'd do the s-c-p things in a branch [11:29] ok [11:29] pitti: that is fine with me [11:29] danke [11:44] update-notifier (0.101ubuntu1) UNRELEASEDmaverick; urgency=low [11:44] mvo: I'll fix that in bzr, ok? since 1.01ubuntu1 is released [11:44] pitti: Were you planning to upload a new apport with the attach_drm_info fixes soon? [11:45] RAOF: would tomorrow be ok? still busy with urgent OEM stuff today [11:45] RAOF: if it's hurting you, please just ask someone to cherrypick and upload, I'll deal with fixing bzr etc. later on [11:46] I'd just like to wait on that before uploading the apport hook which depends on it. It's only hurting in that our bug quality is slightly less than what it could be. [11:51] i wish it was possible to specify which folders (from evo) are monitored in the indicator applet === MacSlow is now known as MacSlow|lunch [11:57] pitti: if you have some free slots, I have 3 sru waiting for some love: light-themes, evince and brasero [11:58] no hurry though :) [11:58] didrocks: will have to wait until tomorrow, I'm afraid (or bug jdong/slangasek) [11:58] pitti: don't be afraid, tomorrow is fine! thanks [12:05] didrocks, pitti: I tried pinging them on #ubuntu-devel, not sure if that will change anything but worth trying to balance a bit the sru load rather than always having pitti and cjwatson having to unblock those [12:05] seb128: sure, thanks [12:12] seb128: do you have a /dev/lp0 or /dev/usblp0 of some sort with the printer turned on? [12:12] seb128: would you mind giving me an "udevadm info --attribute-walk --name=usblp0" output for that? [12:13] pitti, /dev/usblp0 [12:13] ok [12:13] seb128: so I figure you have the "usblp" module loaded [12:16] mpt: hi, I just read your test-case in the SoftwareUpdateHandling page [12:16] hi mvo [12:16] pitti, usblp loaded yes [12:17] pitti, http://paste.ubuntu.com/454399/ [12:17] purrfect, merci! [12:17] mpt: I'm not sure this is really what we want, it seems to me that if the user is presented with the updates he should see them all, no ? [12:18] pitti, you're welcome [12:18] mpt: I mean, I can see why you put it there, but it feels a bit odd, if he wants to apply them all *right now* why shouldn't he be able to? [12:18] mpt: or do you want to have different behvaior for auto-open and manual open? that would make it consistent again, but is IMO rather confusing [12:19] oh, hm [12:19] mvo, the reason it should show only non-security updates is that the security updates should already be installing in the background [12:20] (faketime is a good tool for these kinds of tests btw) [12:20] or have finished installing, even [12:20] rodrigo_, dobey: "ubuntu one disabled" displayed in every nautilus directory, wth? [12:20] seb128, talk to the design team :) [12:20] rodrigo_, dobey: I know it's not enabled, I don't need it to be displayed everywhere [12:20] rodrigo_, who in the design team? [12:20] seb128, we are adding a hidden gconf setting to not display it [12:21] seb128, johnlea [12:21] that seems rather pointless to me? [12:21] seb128, well, the label is just informational, the bar is there to be able to enable it [12:21] why would someone ever enable that gconf key to see the message? [12:21] mpt: right, but that is not necessarily the case that security-updates are installing aleady, it depends on when the background cron script runs [12:21] seb128, don't you see the 'enable' button? [12:21] pitti, the gconf key would be to hide the location bar [12:21] mpt: so I guess it would be better if u-m would monitor changes and remove them on-the-fly once they are installed [12:22] mvo, ok. So how can we fix the test case to test that the security updates *have* installed in the background? [12:22] seb128, if I were the one to decide, I would have patched nautilus-share to support u1, and have all this as part of the 'Share' tab in nautilus properties [12:22] rodrigo_, I do, but I don't want to enable it! [12:23] seb128, yeah, I understand you :) [12:23] mpt: running sudo /etc/cron.daily maually to ensure it is run and has finished should archive that [12:23] it does make sense to have this bar in the ubuntu one folder; but everywhere seems to be a bug [12:23] pitti, you can sync any folder now... [12:23] but the design is just broken [12:23] pitti, yes, you can enable any folder for sync to u1 [12:23] silly question: can gio and the gio async stuff not be used for uid == 0 ? for some reaosn I get "operation not supported" when the same code works perfectly fine as user [12:23] well, but that doesn't mean that I'm going to [12:24] pitti, right [12:24] mvo, ok, and how should faketime be used in that test case? [12:24] rodrigo_, thanks, I'm going to talk to johnlea [12:25] seb128, do you have a u1 account? [12:25] mpt: hold on a sec, I contruct a commandline for tihs [12:25] seb128, because there's still no code to check that, and not show the bar at all if you don't [12:25] rodrigo_, I think I have yes [12:25] ah, ok, then you'll have to see it everywhere :D [12:26] mpt: faketime -f '+1d' update-manager should work [12:26] well, not in folders out of your home, that's also a bug that we need to fix [12:26] mpt: eh, nonsesnse, sorry [12:26] mvo, update-notifier instead? [12:27] mpt: yes, but that needs to be killed before, so killall update-notifier; faketime -f +1d update-notifier [12:27] mvo: faketime? is that an ld_preload which alters the result of gettimeofday()? [12:27] pitti: yes, pretty cool [12:27] neat [12:28] mvo, ok, and how do you tell that the background updates have finished? Does cron.daily wait for unattended-upgrades to exit? [12:28] mpt: hrm, sitll not right: NO_FAKE_STAT=1 faketime -f +1d update-notifier (the NO_FAKE_STAT is important [12:28] mpt: yes [12:29] mvo, so: killall update-notifier && NO_FAKE_STAT=1 faketime -f +1d update-notifier [12:29] mvo: I want a fakeseb128 script! [12:29] pitti: why do you want the fake when you can have the real one? :) [12:29] cause I don't have sudo on seb128 :) [12:29] mpt yes [12:30] heh [12:30] #!/bin/bash [12:30] echo "is gtk bug" [12:30] "iz" [12:30] mpt: but right, that'll get us 90% of the way [12:31] See, I can't even write a two-line script without at least one bug [12:31] we've got no chance :) [12:32] $ apt-get source seb128 [12:32] E: Unable to find a source package for seb128 [12:32] mvo, so when you run that, update-manager should open immediately? [12:32] lifeless: I'm afraid we don't, right [12:33] seb128, btw, if I understood correctly the notify-osd page, for a notification coming from a daemon (that is, no window to show the notification in), the morphing window/alert box should be shown in the background, behind all open windows, right? [12:34] mpt, ^^ [12:34]