[00:14] james_w - i've pushed a change to bzr for the gnome-desktop issue now, incorporating your change too [00:14] i made a small change though. i removed this bit: [00:14] + *error = NULL; [00:14] as that leaks the GError if it was set [00:15] because the error is just discarded here, I don't pass it to the previous function at all anymore [01:48] TheMuso, you there? [01:53] i miss seb :( [01:59] desrt, :) [01:59] desrt, I am your Canonical appointed Seb-substitute. How can I help ;) [02:00] TheMuso, can you release lp:~ubuntu-desktop/xdg-utils/ubuntu? Needs main privileges [02:01] robert_ancell: a hug? [02:01] robert_ancell: What does releasing involve? [02:01] oh upload it? [02:01] TheMuso, er, uploading. Sponsoring. Whatever the jargon is :) [02:02] fixes attachments and thunderbird [02:02] (patches taken from upstream [02:02] ) [02:02] robert_ancell: Righto will get to it in a sec [02:02] no rush [02:02] * robert_ancell hugs desrt [02:02] robert_ancell: that was most excellent. thank you. [02:06] TheMuso, oh, also any feedback about simple-scan and a11y would be appreciated. I'm thinking OCR support could be very useful. [02:06] robert_ancell: Ok haven't looked at it yet, but I'll give it a try when I next need to scan something. [02:07] TheMuso, are there any tools that scan and read out the contents? [02:07] robert_ancell: mago/ldtp, accerciser to name a couple. [02:07] robert_ancell: that is if you mean a11y testing [02:08] TheMuso, I mean read out the contents of the scanned page [02:08] robert_ancell: oh right, none that I know of that do that. [02:08] TheMuso, I do use those tools but it's often hard to tell if ticking all those boxes really makes it accessible [02:08] * robert_ancell files a new bug [02:19] robert_ancell: ok looking at xdg-utils now. [02:30] robert_ancell: uploaded [02:30] TheMuso, thanks [02:30] np [03:07] RAOF, hey, thanks for the launchpad-integration work === nick_ is now known as Guest46551 [07:04] <^arky^> hi [07:05] <^arky^> Wonder what happened to shutdown menu under system its an important a11y bug 510775 [07:05] Launchpad bug 510775 in indicator-applet "indicator applet removes shutdown menu option" [Low,Invalid] https://launchpad.net/bugs/510775 [07:15] Good morning [07:16] asac: langpack machine name> macquarie [07:16] kklimonda: they are hacked somewhat, like dropping the rsyslog dd process; but not too much [07:38] good morning [07:38] bonjour baptistemm [07:40] hi pitti [08:23] good morning [08:25] bonjour didrocks [08:25] hey pitti, did you have a good week-end? [08:25] oh yes, I did; it was great for relaxing [08:26] how about you? [08:30] pitti: I connected to IRC and got a lot of questions and feedbacks (more details on that later), so not really resting finally :) [08:30] ah, error [08:31] after not having had a weekend last week, I really needed this one [08:32] didrocks, sorry :) [08:32] hello :) [08:33] baptistemm: not particularly your fault ;) I had some ping on quickly, anjuta, and une too ;) [08:33] salut seb128 [08:33] bonjour seb128 [08:35] salut seb128 [08:35] hey didrocks baptistemm pitti [08:35] didrocks, never IRC during the weekend ;-) [08:36] seb128: right, I guess I have to change my habits :) [08:37] or change your nick during the weekend [08:37] :) [08:38] hehe /nick didrocks_hidden === geser_ is now known as geser [08:41] pitti, what is the difference between devicekit-disk and udisk ? [08:42] baptistemm: it was by and large a rename, but udisks also grew some more features [08:43] okay [08:43] the api has changed ? [08:43] just uploaded usb-creator [08:43] I'll port dell-recovery now, then I can remove devicekit-disks from the archive [08:44] baptistemm: yes, due to the renaming; but not a lot besides that [08:44] i. e. o.fd.DeviceKit.Disks -> o.fd.UDisks [08:44] likewise with d-bus object paths, and so on [08:48] usb-creator will need porting too I guess? [08:48] 09:43:36 pitti | just uploaded usb-creator [08:48] oh [08:49] sorry [08:49] haven't had enough coffee yet [08:49] np :) [08:49] Keybuk: good morning [08:49] Keybuk: how was your weekend? [08:49] relaxing thanks, yours? [08:50] for me, too; went for some skiing, visiting family, grandparents in hospital, and got some housecleaning done [08:50] oh, and some episodes of Al Bundy :) [08:53] heh, nice [08:53] most of mine was spent with friends [09:10] restarting session to test something, brb [09:22] baptistemm: bg change commited [09:23] youpi [09:24] Ihaz sharp Eyze [09:24] baptistemm: you have to say "chouette" here, it's rick's rule :) ("swcheet" is also accepted ;)) [09:24] Chouette !!! [09:24] good morning everyone [09:24] baptistemm: heh, you will tell me once sponsored and installed [09:24] so what did you do? [09:24] didrocks, so not before tonight [09:25] baptistemm: put the quality to 100% for jpeg, the image have approximately the same size [09:26] chrisccoulson: work late again? (see a commit at 00:10 on libgnome-desktop) :) [09:26] hey didrocks [09:26] that's not late for me ;) [09:27] chrisccoulson: sure, relatively speaking… ;) [09:28] didrocks: did you have a good weekend? [09:29] chrisccoulson: a lot of IRC ping as you saw on Saturday for instance (but the same on Sunday). So, not really resting as I'm bugged myself during whole the week-end about the session issue :) [09:29] but it was still a week-end :) [09:29] chrisccoulson: and you? [09:30] i had quite a relaxing weekend. i had plans to do lots of work, but we had family round in the end, and I didn't really start anything until yesterday evening [09:30] didrocks, format name is always "jpeg"?, it couldn't be jpg for instance ? [09:31] baptistemm: looking at the format doc, it seems not [09:31] okay, sorry for asking [09:31] heh, i see pitti is busy this morning moving dk-disks bug tasks over to udisks ;) [09:31] yeah, good oppportunity to review dk-disks bugs and catch up on triaging [09:32] good morning [09:32] baptistemm: no pb. If we see some special case, we will be able to change that easily :) [09:32] good morning pitti :) [09:32] how are you? [09:32] chrisccoulson: I'm great, thanks; had a nice and relaxing weekend. how are yuo? [09:33] yeah, quite relaxing too. we left our daughter with her grandparents on saturday evening, and went out for a meal [09:33] which was a nice break! [09:33] bad parents !!!! [09:33] :) [09:33] lol [09:34] we only left her for an hour or so [09:37] pitti - i just saw you left a comment on the g-d-u bug that superm sent upstream at the weekend [09:37] *nod* [09:37] i thought we'd fixed this issue [09:37] but the new fix is definately wrong [09:38] I faintly remembered that we discussed that before, indeed [09:38] but I wasn't sure whether it was the same problem [09:38] pitti - the issue we identified was related to the timeout of the dbus calls being shorter than expected [09:39] the issue that superm has reported is fundamentally the same issue [09:39] (the first dbus call fails when activating the daemon) [09:40] pitti - i'm wondering if there isn't another race here with activated services. udisks claims the org.freedesktop.udisks service name (or whatever it is called now) when it loads, but it does this before registering the object we're trying to communicate with [09:41] i wonder if that can make the first call fail sometimes [09:41] oh, is it even possible to do that the other way round? (register, then claim) [09:42] pitti - i can't see why not. you'd still have a connection to the bus (but it would be a number) [09:42] then you just change the name when you're ready [09:42] perhaps james_w has some insight here [09:43] james_w - FYI - bug 521481 (this issue keeps cropping up again) [09:43] Launchpad bug 521481 in gnome-disk-utility "gvfs-gdu-volume-monitor crashed with SIGSEGV in gdu_pool_get_presentables()" [Medium,Fix released] https://launchpad.net/bugs/521481 [09:43] pitti - do you want me to reopen this bug? [09:44] chrisccoulson: yes, I think we should [09:45] pitti - done [09:52] didrocks - you bumped the gnome-desktop version number in bzr to 2.29.90-0ubuntu4 [09:52] For thoses interested, I packaged devhelp 2.29.90, I just need sponsoring [09:52] there's no need to do that is there? [09:52] chrisccoulson: grrr, I used dch -a, sorry, let me fix that [09:52] (as 2.29.90-0ubuntu3 hasn't been released yet) [09:52] ah daniel did it [09:52] heh ;) [09:52] thanks:) [09:52] chrisccoulson: sorry, twice in a week ;) [09:52] chrisccoulson: fixed [09:53] thanks [10:04] pitti, bug #453605 [10:04] Launchpad bug 453605 in nautilus "Make default mount umasks configurable" [Wishlist,Triaged] https://launchpad.net/bugs/453605 [10:04] pitti, this sort of option used to be in gnome-mount [10:04] right, g-m had a nautilus extension [10:05] pitti, I don't nautilus is the right component for such policy, should that be somewhere in the udisk stack? [10:05] or g-d-u [10:05] no, it needs to be in GNOME, and nautilus does the automounting [10:05] (which I'm not happy about, but that's the way things are ATM) [10:05] well nautilus doesn't specify mount options does it [10:06] right, that's the wishlist :) [10:06] I don't think nautilus should [10:06] udisk should have some config file it reads [10:06] the way gnome-mount was reading gconf keys [10:06] no? [10:06] it would seem logic to not have those options restricted to the caller [10:06] udisks doesn't have policy, or can have per-user config [10:06] we could have nautilus or gdu read gconf values [10:07] well, what does call mount in the stack [10:07] and why could that read a config? [10:07] I'm sure there is non nautilus users who want to tweak mount options too [10:08] well, you can with udisks and gdu [10:08] gdu_device_op_filesystem_mount() takes options, and so does udisks --mount [10:09] in theory we could add gconf reading to gdu [10:09] that would make it a bit more awkward to XFCE and friends (due to the new dependency), but I don't care that much about this [10:09] let's say it's a wishlist for next cycle [10:10] by which time we will have gsettings in glib [10:10] and we can use that rather than gconf [10:10] pitti, thanks! [10:10] right, just a wishlist bug; can be entirely ignored [10:10] yeah, will do that [10:10] I was just curious about the design [10:10] ah, then we could add configurability to gdu indeed [10:10] seems gnome-mount flexibility was better ;-) [10:10] since it'll need glib either way [10:11] yeah, it was indeed [10:12] hey seb128 [10:12] hey chrisccoulson [10:12] how are you today? [10:13] chrisccoulson, good! you? [10:14] seb128 - yeah, i'm ok thanks. i'm getting a bit tired now though, as I've ran out of coffee at work [10:15] no coffee on monday morning is not a nice week start ;-) [10:15] it's not ;) [10:15] i have to wait until lunchtime before i can get some more [10:16] chrisccoulson, you're a coffeelcoholic ? [10:17] baptistemm: yeah, i am [10:17] it will probably do me good to have a morning without though [11:02] For those working on Moblin, http://meego.com/ --> « Moblin and Maemo are merging » [11:02] you can say bye bye to GTK+ development on mobile platform ... [11:26] cassidy, do you look at empathy bug in launchpad ? [11:26] baptistemm, rarely. I already receive too much upstream bugs :\ [11:26] okay [11:26] So I need to forward my crasher [11:32] baptistemm: if it's not due to one of our patches ;) [11:32] we'll see :-/ [11:44] baptistemm, how does it crash? [11:45] baptistemm: is bug 520603 the one? [11:46] Launchpad bug 520603 in empathy "Empathy crashed after trying to send a file to a contact from nautilus" [Medium,New] https://launchpad.net/bugs/520603 [11:46] cassidy: ^^ [11:47] * chrisccoulson wishes it was lunchtime already [11:48] chrisccoulson, this one should have been fixed this week-end [11:48] chrisccoulson, you can close it as a dup of https://bugzilla.gnome.org/show_bug.cgi?id=609843 [11:48] Gnome bug 609843 in Chat "Crashed when I closed one tab and hence switched to another tab for a disconnected account" [Major,Resolved: fixed] [11:49] cassidy - thanks [11:53] pitti: can I get some feedback on that before implementing it: https://wiki.ubuntu.com/DesktopTeam/Specs/Lucid/DesktopUNESession#gnome-panel ? [11:59] seb128: same punition (as you were disconnected). When you have some time, can I get some feedback on that before implementing it: https://wiki.ubuntu.com/DesktopTeam/Specs/Lucid/DesktopUNESession#gnome-panel please? [11:59] same punition than who, where? [11:59] punishment even :) [11:59] seb128: I asked pitti the same thing [11:59] :) [11:59] looking [11:59] thanks [12:01] I like none of those [12:01] did you ask vuntz for his opinion? [12:02] chrisccoulson, yeah thank, I've already create one bug upstream, I'll close it [12:02] vuntz told me to do the #3 when we discussed that, but I'm afraid it can break a lot of things (but I didn't show him this) [12:02] vuntz: https://wiki.ubuntu.com/DesktopTeam/Specs/Lucid/DesktopUNESession#gnome-panel, any idea on how to fix this? [12:04] vuntz: because even, if we got fixed by #3, the user won't be able to add a new panel still as it would be mandatory and not depends on session (or if we make it default, if the user is adding/changing anything, he will get the same panels in UNE as well) === kermiac is now known as kermiac_ [12:13] didrocks: will look after lunch [12:14] pitti: thanks, enjoy :) === MacSlow is now known as MacSlow|lunch [12:55] asac, I sent a mail to linux-bluetooth mailing list about the API break, Waiting for an answer now [12:56] thx [12:56] I don't know would what be the better fix, bump soname or restore API [12:57] I don't know what use the api which was removed [12:57] restore API i would hope [12:57] just wait for upstream ,) [12:57] their call imo [12:57] fedora pushed the package anyway [12:57] they dont have symbol tracking, do they? [12:57] they probably didnt notice [12:57] I wonder [12:58] hey Bastien told me he didn't noticed [12:58] right [12:58] thats what i suspected :-Ü [12:58] :-P [12:58] or rather he didn't told me he noticed :) [13:08] pitti: after having a look of g-s-d code about triggering the wallpaper recache on update for the WI "write a small helper program to refresh bg and trigger caching, and add it to gnome-desktop postinst", I noted that it's watching the wallpaper file. If you change it, you trigger g-s-d to redraw to background, and so, get the new wallpaper cached. [13:09] pitti: consequently, I wonder if it's useful for people not having the background in a package. Do we still want to cache their wallpaper on update, or just wait for next reboot to do it for us? [13:10] didrocks, next reboot is fine [13:11] didrocks, the profiling etc will only update on first boot after upgrade [13:11] didrocks, I was away for lunch looking at your une question again, but I think we should do what vuntz suggests [13:11] didrocks, the spec is not really clear about what you aim to fix though [13:12] seb128: the issue is even if we fix that, we prevent people from adding another panel (as toplevel_id_list will be mandatory), right? [13:12] didrocks: hi [13:12] good afternoon everyone [13:12] didrocks: yes, I think we want to cache it either way [13:12] chrisccoulson, good afternoon to you too [13:12] didrocks: but touching it during dist-upgrade doesn't work, since at that time the old g-s-d is still running [13:12] good afternoon chrisccoulson [13:13] pitti: oh right, forgot about that [13:13] didrocks, it doesn't make clean either what was the drawbacks from the IRC discussion nor the bug reports [13:13] didrocks: and doing it on first boot will just miss the chance of being covered by ureadahead [13:13] didrocks: (same problem why we need an ubiquity hook) [13:13] seb128: "not editing the panel?" [13:13] seb128: I can point the IRC discussion though [13:13] didrocks, is that something you want? [13:13] let me find it [13:14] being able to edit the panel or not [13:14] didrocks, I'm not really interested to read IRC backlogs if you could summarize the concern rather... [13:14] seb128: just that we already have bug report about that and bad feedbacks before alpha3 about preventing that [13:14] seb128: in a nutshell, the feedback was "UNE sucks if I can't edit the panel" [13:15] and other constructive terms like that :) [13:15] you can't edit the panel at all with mandatory keys? [13:15] (took one hour of discussion on a Saturday, not always pleasant) [13:15] seb128: right, you can't [13:15] or just not move the mandatory components? [13:15] you can't add/remove/move [13:15] (and can't add any other panel too) [13:16] pitti: ok, so, make sense for the caching thing. Not as easy as I guessed at first glance [13:17] didrocks, do you have a statement of what the une session is meant to be? [13:17] I'm being confuse about the purpose [13:17] is that something to make easy to try une [13:17] is that something which should allow full experience on both sessions [13:18] or rather something to have a feel of what une is? [13:18] seb128: that's what I mostly answered on the bug reports "if you want to edit panels, use GNOME session now available" [13:18] ie should config changes goes over to gnome session and back [13:18] you might want to provide a script to turn the GNOME session in une [13:19] seb128: hence my first sentence. Do we want to state "the default panel is the UNE experience and we don't want user to change it" [13:19] so users who want to use une can run that and will get the gnome-panel configured etc [13:19] and keep using their normal session [13:19] which will be an une one [13:19] the script would set gnome-panel user config to une layout [13:19] and change some autostart .config desktop files etc [13:19] and say that the une session is a locked une demo [13:20] those who really want to use une daily can turn their default session to an une ... [13:21] seb128: hum, difficult as UNE as dedicated applets (home-swichter, window-picker-applets) that you don't want in your normal GNOME session. Also, how determine that the user just wants one panel, and so on… [13:21] ? [13:21] there is no GNOME session in what I suggest [13:21] the script would configure your GNOME as an UNE session [13:22] you wouldn't have a GNOME session [13:22] that's for those who want to use UNE and not GNOME stock [13:22] oh ok [13:22] you would have the UNE locked session for testing [13:22] some kind of "revert to previous state", only having one session [13:22] and the option to run this convert script [13:22] to make your GNOME session match UNE [13:23] in which you would have no mandatory etc [13:23] the issue is that we would write/cp some files outside of any packaging management [13:24] well that's user config [13:24] it's not an issue to provide configuration tools [13:24] I see 2 options [13:24] I'm thinking about the gconf keys in /usr/share/gconf/ [13:24] either do what I just described [13:25] or do 2 independant sessions using option3 [13:25] in which case they would be really independant [13:25] ie not share clock locations etc [13:25] that would be by design [13:25] since they are independant === MacSlow|lunch is now known as MacSlow [13:27] hum, right, but that can be quite short to achieve before FF (as I guess they asked that from gconf and there is no "other user session" gconf). You lost all your apps config and such [13:27] as it's asking gconf for the right value [13:28] what is "that"? [13:28] in "they asked that" [13:29] seb128: configuring the panel [13:30] didrocks, I'm not sure I follow you there [13:30] I would say check with rick and,or pitti what you want the experience to be first, then decide on what to do... [13:30] ie do you want the session to be locked or not [13:30] do you want it to be a demo or something users can run daily and customize [13:31] should the config changes go to the GNOME session too or not, and the other way around too [13:31] seb128: that's why I wrote this on the spec, to share with you, pitti (and rick when he will be there) [13:32] I don't feel I know enough about the goal of this work to have an opinion [13:32] I can give technical opinion on how to achieve one of the goals [13:32] but I've not clue about what the purpose of this session should be [13:32] seb128: the thing is that the session is not only a "UNE demo", it's UNE (there is no difference with UNE stock install) [13:33] seb128: the fourth solution would seem to be the best experience-wise, but if as you say that's hard to implement, nevermind [13:33] sorry, didrocks ^ [13:34] please don't set up a parallel gconf tree or something like that [13:34] didrocks, UNE stock install use mandatory settigns too? [13:34] pitti, why not? [13:34] pitti, it would not be a dup gconf config but just gnome-panel and gnome-panel-une [13:35] pitti, ie namespace some keys and patch gnome-panel to read the namespaced version [13:35] seb128: right, a new gconf dir /apps/gnome-panel-une/ is fine for me [13:35] but not a ~/.gconf-une/ [13:35] pitti: the 4th solution will only force to have one panel and what applets you have in your session (as the mandatory key are still needed for /apps/panel/general/{applet_id_list,object_id_list,toplevel_id_list} [13:35] seb128: I agree [13:35] didrocks: fundamentally I'd consider the UNE panel and GNOME panel as two separate entities wrt. experience and configuration [13:35] pitti: this is the second solution in fact (something like /apps/gnome-panel-session), more generic [13:35] and not try to "merge" their configuration space [13:36] if it's ok for you and seb128, I can do that (instead on working on keys, just /apps/panel-{session} [13:36] didrocks: /apps/panel/general/toplevel_id_list/toplevel_id-une is a bit too interleaved for my taste, but I could live with it [13:37] pitti: right, I like more your idea, more generic :) [13:37] I'd actually prefer /apps/panel-${session}/general/... [13:37] seb128: ? ^ [13:37] I'm not really sure [13:37] I think the namespace one is the best way to achieve what you want [13:38] but that will only unsync the gnome-panel config changes [13:38] not sure if others should be considered [13:38] I've the feeling this is turning to something not easy to get working and bug prone too [13:38] which I would like to avoid for the lts [13:39] my favorite option right now would be what I described before [13:39] declared UNE a demo locked session [13:39] and let people who want to use a customize UNE to turn their GNOME into one [13:39] I think the UNE session is not worth destabilizing the default desktop for a lts [13:39] seb128: but there is no "demo" and the default UNE will be the same (there is no difference in installing ubuntu-desktop and then UNE or UNE and then ubuntu-desktop) [13:40] just to ensure you know that, it seems you think that's a seperate functionnality :) [13:40] didrocks: I just don't understand how this is suddenly a new problem? previous UNE releases certainly didn't behave any different? [13:41] didrocks, I don't see why you couldn't dump the gnome-panel une profile on UNE install [13:42] pitti: in fact, previous version were using the gnome session, so you weren't able to have both sessions [13:42] didrocks, well right not you could have UNE without mandatory use by writting the correct gnome-panel profile [13:42] seb128: installing gnome session on une was also on the spec. If we copy that, we can't have the gnome session back [13:42] didrocks: oh, it's because in normal ubuntu you can now install ubuntu-netbook to get the netbook-launcher session in gdm as well, and that would lock your main GNOME panel? [13:42] not -> now [13:43] pitti: right, all the tree is copied on install [13:43] seb128: on karmic? this is because you can't run ubuntu-desktop as well [13:45] didrocks, no, I'm not speaking about karmic [13:45] but I think what I'm suggesting is to get on lucid what we had on karmic [13:45] + current hack which give you a locked UNE session [13:45] or mandatory gnome-panel UNE config [13:46] or whatever you want to call it [13:46] seb128: the first solution remove a part of the spec "having UNE and install ubuntu-desktop session on it" [13:46] (if we do that by default on une installation) [13:47] but well, keeping it mandatory can be a good thing, it was really difficult to add new applets to previous UNR panel [13:48] didrocks: how can you keep the UNE panel mandatory while still leaving the GNOME one alone? [13:49] didrocks: we have some magic to include (or not) /usr/share/gconf/une.mandatory.path on a per-session basis? [13:49] pitti: you add to gconf a path depending on gdm session which is /var/lib/gconf/.{mandatory,default} [13:49] ah, I see [13:50] didrocks: so, if locking the GNOME panel when ubuntu-netbook is installed isn't an issue, just lock it and say "that's the UNE panel, dammit" [13:51] pitti, user gnome-panel is not locked right, it's only the UNE one which use mandatory settings [13:51] and yeah, +1 from me to say that [13:51] pitti: that's what we have from the start [13:51] and how I imagined the thing [13:51] + maybe provide a small wrapper which dump UNE config on your user [13:51] to turn your GNOME session in une [13:51] well, I just had a lot of complaints, even in a alpha :) [13:52] you always have people to complain [13:52] seb128: right, for those user, we can do that [13:52] "wontfix" then [13:53] ok, and about sharing the same time applet? do we keep two seperate? [13:53] whatever is easier [13:53] I want to avoid hacks, especially those who could create bugs in the GNOME session [13:54] ok, I'll speak about that with rick just to confirm that's the way we go and if we say "this is the une panel", nothing to touch so :) [13:55] good [13:55] and tell users who really use UNE to turn their default session to an UNE one [13:55]