=== marcos_ is now known as Daekdroom === m_conley is now known as m_conley_away === Amaranth_ is now known as Amaranth === achiang` is now known as achiang [07:58] good morning [08:05] hey there everybody [08:17] good morning [08:37] Does anyone have working Gtk appmenus anymore in unity-2d-panel? Qt appmenus still seem to work. === ogra_ is now known as ogra [10:18] murrayc: ping [10:20] smspillaz: pong [10:23] murrayc: seems that glibmm destroys all of your sources once the mainloop exits. is that the intended behaviour? [10:24] since I store them in PrivateScreen, and when PrivateScreen gets destroyed, it calls ::unreference () on the sources (Glib::RefPtr <> does this) and then segfaults because the sources are already gone [10:24] They would be destroyed when the C object is destroyed, meaning when their refcount gets to 0. [10:25] I doubt that anything explicitly destroys them before the refcount gets to 0. [10:25] valgrind would tell you. [10:26] murrayc: valgrind says they're being destroyed on g_main_context_dispatch (I assume when the mainloop has quit) [10:27] http://paste.ubuntu.com/573388/' [10:27] erm [10:27] http://paste.ubuntu.com/573388/ [10:28] That's missing some detail. Maybe you dont' have the debug packages installed. [10:28] I don't think that really tells you what is happening. [10:29] murrayc: I know :) Just I know that something weird is up :) [10:29] nevertheless, I was going to get the -dbg stuff, getting it now [10:30] It might just be a missing reference() somewhere, if we are lucky. [10:31] murrayc: yeah. I first tried to ->reference () when I create all the sources, though, that doesn't help === API is now known as Guest87772 [10:31] smspillaz: It's also possible that it's not the source itself, but something of yours that uses it, that's being double-deleted. [10:32] If it doesn't become clear then a simple test case is the logical debug step. [10:32] murrayc: indeed [10:33] murrayc: I checked for wrong uses of ::reset () and ::clear () too ... nothing there ... === Guest87772 is now known as apinheiro [11:40] apinheiro: ping? [11:40] dbarth, pong? [12:07] njpatel, from a PanelIndicatorObjectView, how do I get the indicator id from it? GetName is protected, and can't find anything else [12:10] rodrigo_, _entry->GetId () [12:10] rodrigo_, _proxy->GetId () [12:10] ok [12:11] it's public (_proxy) [12:11] _proxy on the PanelIndicatorObjectView? it's protected [12:12] ah, from the EntryView [12:14] right === MacSlow is now known as MacSlow|lunch [12:29] May I ask indicator applet developers here to review a merge proposal for bug 724917 - it's highly annoying and a fix is available for a couple of days [12:29] Launchpad bug 724917 in libappindicator (Ubuntu) "Importing appindicator from python crashes with ImportError on undefined symbol" [Medium,Triaged] https://launchpad.net/bugs/724917 [12:49] dbarth, in the end, after the last changes on nux and unity, it would be not required that custom script on orca [12:49] I have tested it, and orca properly speech the selection changes [12:50] when you start the key nav mode on Launcher [12:50] rodrigo_ is reviewing that branch [12:50] but, could any other one review it? [12:50] https://code.launchpad.net/~apinheiro/unity/atk-selection/+merge/51440 [12:50] as I also made some changes on the Launcher [12:50] njpatel, ^^ [12:52] apinheiro, yep, i'll ask macslow to review it before release [12:52] njpatel, ok thanks [12:52] MacSlow|lunch, when your back and while your waiting for smspillaz, please review ^ [13:21] apinheiro: ah awesome, so that shows that the changes are all coming together [13:24] njpatel: anything you know of that could be causing the unity launcher to be sliding in and out really fast? [13:24] klattimer, yep, it's a known (and fixed in trunk) bug [13:24] njpatel: how long until it'll be out? [13:24] klattimer, just unity --replace a few times and it'll sort itself out [13:24] hmm, ok [13:24] klattimer, dude *never* ask that question on feature freeze day ;) [13:24] klattimer, later on today [13:24] well, not FF day [13:25] but last-day-before-your-banned-from-uploading-day [13:25] you can upload all the time :) [13:26] it will just hang in the queue until fater milestone release [13:26] ;) [13:26] heh === m_conley_away is now known as m_conley === MacSlow|lunch is now known as MacSlow [13:59] njpatel, apinheiro: ok [13:59] thanks! [14:00] kenvandine, ping [14:00] nmarques, pong [14:01] kenvandine, those 2 patches did wonders and also fixed a glitch on transmission with the indicator [14:01] woot [14:01] kenvandine, but I noticed I still have one problem [14:02] kenvandine, on indicator-me, if you open empathy through indicator-messages, it doesn't allow you to change status on indicator-me, unless you restart the applet (with empathy already opened) or kill the gnome-panel [14:02] humm, not having that problem here [14:03] oh wait [14:04] kenvandine, it happened after the dbusmenu update [14:04] maybe i am [14:04] before that was doing fine :) [14:04] ok, this looks like the same bug we are seeing with menus that aren't sensitive [14:05] i didn't notice because i had done a killall indicator-me-service last night after empathy was running :) [14:06] it's cool, I noticed it by pure luck also [14:06] there are a couple places where this is happening [14:06] this might be an easier place for me to debug it... will look [14:06] kenvandine, mind of a question regarding appmenu? [14:06] was waiting for tedg to get back from scale [14:06] sure [14:07] kenvandine, some applications if the window is destroyed the appmenu indicator doesn't display the menu anymore [14:07] I've noticed this on 2 apps [14:07] empathy (contact list is destroyed) and if then is opened it doesn't populate the menu [14:07] * kenvandine tries [14:07] and only the File/close entry is displayes [14:08] empathy has a similar behavior [14:08] ups, I mean Banshee [14:08] yeah, that has been happening for a while [14:08] i had only seen it in empathy [14:08] well I'm running Banshee 1.8.0 which is the one shipped on openSUSE 11.4 [14:08] * aruiz ponders the idea of buying boycottnvidia.com [14:08] there is a bug about it, one of the many things we'll get to now that we are past feature freeze :) [14:09] aruiz, awesome :) [14:09] aruiz, :) [14:09] aruiz, they don't care :) [14:09] aruiz, i boycotted them years ago, worked great for me [14:09] :) [14:09] nmarques: I'll make them care [14:09] aruiz, I still love them for their robust GL-driver [14:09] aruiz, I had 4 different motherboard on one of my laptops with a G86 running over 82ºC [14:10] intel graphics might not be as awesome... but it is generally pretty painless :) [14:10] aruiz, they never cared ;) [14:10] aruiz, I never bought nvidia again [14:10] * aruiz misses the 3Dfx days [14:11] aruiz, amen to that brother! [14:11] aruiz, hey they are part of nvidia now [14:11] MacSlow: I know [14:11] RIP [14:12] kenvandine, I'll hold a bit, once the fixes are out, I'm going to freeze this for 11.4 and start hasting on passing everything to GNOME3 and retake unity (hopefully Sam will merge the glib branch meanwhile) [14:12] kenvandine, if you guys also need help with testing, let me know, I can for sure help with that either on Ubuntu or *SUSE [14:12] i think he did and actually did a release [14:12] at least i know they were working on that last week [14:13] nmarques, we always want testing :) [14:13] kenvandine, I've had very good feedback from this [14:14] nmarques, great [14:32] apinheiro, -> https://code.launchpad.net/~apinheiro/unity/atk-selection/+merge/51440 [14:33] MacSlow, see my last comment [14:33] I was able to solve that crashes [14:33] and I updated the branch since then [14:34] apinheiro, ups... didn't see that :) [14:34] just pulled and didn't further look at the commit-history :) [14:35] yeah, I know that normally you don't update the branch until the review of others ;) [14:35] apinheiro, ok [14:35] apinheiro, approved [14:35] MacSlow, ok, thanks [14:36] I will merge it, although I guess that I'm out of time now [14:36] I guess that the deadline was 30 minutes ago === _LibertyZero is now known as LibertyZero [14:37] apinheiro, ask njpatel if it's still good to go in [14:37] njpatel, you here? [14:38] hey [14:38] apinheiro, yes, please merge it [14:38] njpatel, ok === dbarth is now known as dbarth-otp [14:47] kamstrup, is it a bug that setting an emblem on a launcher icon changes the background color? [14:49] kenvandine: i was just wondering that this morning... [14:49] kenvandine: I think it maybe happens because of unity's intelligent background color choosing [14:49] kenvandine: it sees the icon change and think's it should recalc the bg color, but I think it's a bug [14:50] kenvandine: although I think DBO will be the authoritative person to ask here [14:51] ok, i will file a bug either way to document it [14:51] it has to be a bug :) [14:51] hiding the emblem doesn't reset the color === apinheiro is now known as apinheiro_afk [14:57] kamstrup, i filed it against libunity, since that is where the problem is exposed... but i am sure the bug is actually in unity [14:58] njpatel, can I get a 2nd review of this: https://code.launchpad.net/~rodrigo-moya/unity/ref-state-set-for-panel-service/+merge/50955 please? [14:58] njpatel, also, have you already done today's release? [14:58] * kenvandine is surprised how few bugs there are against libunity [14:58] guys, what would be the best place to provide a 'Wishlist' 'bug report' regarding the whole indicators (maybe not indicator-appmenu, but all the others) ? [15:01] rodrigo_, I haven't done today's release yet [15:02] rodrigo_, I can't review right now, will find someone to do it, hold up [15:02] njpatel, ok, so please review that branch and this other one: https://code.launchpad.net/~rodrigo-moya/unity/accessible-name-for-indicators/+merge/51120 so that they are included in the release, please? [15:02] njpatel, ok [15:03] guys, while you are at it, could you also review this merge proposal? https://code.launchpad.net/~artfwo/libappindicator/fix-for-bug-724917/+merge/51411 [15:22] klattimer, are you aware of any problems with indicator-datetime not displaying all appointments? === dbarth-otp is now known as dbarth [15:22] kenvandine: not to my knowedge [15:23] i have appointments on my calendar for almost everyday this week [15:23] kenvandine: it updates every 5 minutes now, and updates on start [15:23] but only a couple of them are displayed [15:23] rather than about-to-show [15:23] none of these are new [15:23] hmm [15:23] have you just rebooted? [15:24] no [15:24] i restarted the service [15:24] and it didn't make a difference [15:24] kenvandine: if you could watch the output from the indicator-service [15:24] i see lots of these [15:24] (process:4109): Indicator-Datetime-DEBUG: Appointment timezone: (null) [15:24] that should tell what's happening [15:24] i never see debug output for the appointments that aren't being displayed [15:24] and i see dupes of everything except the last one [15:25] (process:4109): GLib-GObject-CRITICAL **: g_object_unref: assertion `G_IS_OBJECT (object)' failed [15:25] and that right after it says it is adding it [15:25] but that last one is from this coming friday [15:25] One Time Pad, best cipher there is :) [15:25] i have a bunch before friday that aren't showing up [15:25] kenvandine: do you see any "Summary: title of appointment" [15:26] yeah [15:26] * klattimer hopes he doesn't have to distro patch this :/ [15:26] and i am only seeing those for the ones i am seeing in the menu [15:27] kenvandine: are there _ANY_ appointments showing? [15:27] yes [15:27] 5 [15:27] that's how many you should see [15:27] there is a maximum of 5 displayed at a time in the spec [15:27] ugh... but i am seeing some for friday [15:27] but not any for today! [15:27] they'll change as they pass through their time [15:27] kenvandine: can you work out which? [15:28] it *might* be that your google calendars are lagging a lot [15:28] i see it on the calendar [15:28] and it isn't new [15:28] it is a recurring one every monday [15:28] kenvandine: which source does it come from? [15:28] my work calendar (google) [15:28] (seeing it in the calendar means nothing as eds is strange) [15:28] i see others from that calendar [15:28] later in the week [15:29] hmm [15:29] then that is strange [15:29] could it be timezone related? [15:29] they are getting (null) for the timezone [15:29] kenvandine: unlikely [15:29] ok [15:29] I'm thinking it might be timer related [15:30] and actually the ones not being displayed are from 3 different calendars [15:30] including the default local one [15:30] looks like the timer isn't actually running at all [15:33] kenvandine: can you look through the initialisation debug [15:33] and tell me if you see it failing on google source? [15:33] sure [15:34] also kenvandine is this the most upto date i.e. with the newest libido and indicator-datetime from trunk? [15:34] from natty [15:34] hmm, the timer _is_ running, except for some reason it stopped [15:34] which afaik is the same [15:35] kenvandine: there should be an update coming today actually [15:35] i don't see failures loading a calendar [15:35] hmm [15:35] but... does it aggregate all the events together before sorting them? [15:35] then it's not my code [15:35] chances are something is wrong with either eds or google [15:35] kenvandine: it does, all of them are an aggregate and sorted by time [15:35] perhaps it is showing me 5 from one calendar and truncating? [15:35] ok [15:37] oh... how does it sort? [15:37] by startdate? [15:37] i have lots of recurring appointments [15:37] kenvandine: yeah [15:37] or due [15:37] ah! [15:37] rodrigo_, ping [15:37] MacSlow, pong [15:37] so the ones displayed all have startdates of 2009... [15:37] kenvandine: basically it's sorted by the time which would display in the menu [15:37] kenvandine: that's interesting [15:38] i wonder why :/ [15:38] klattimer, so you need to sort it by a different field [15:38] rodrigo_, there are two branches you've against unity... which do you need review-eyes on still? [15:38] recurring ones must have more data in there for next occurance or something [15:38] * kenvandine files a bug [15:38] MacSlow, https://code.launchpad.net/~rodrigo-moya/unity/ref-state-set-for-panel-service/+merge/50955 and https://code.launchpad.net/~rodrigo-moya/unity/accessible-name-for-indicators/+merge/51120 [15:38] rodrigo_, both then ok [15:42] rodrigo_, approved with small commend -> https://code.launchpad.net/~rodrigo-moya/unity/accessible-name-for-indicators/+merge/51120 [15:43] kenvandine: not sure how this can be fixed [15:43] it'll take a while to look into it [15:43] can you file a bug [15:43] done [15:43] bug 726603 [15:43] MacSlow, ok [15:43] Launchpad bug 726603 in indicator-datetime (Ubuntu) "Appointments displayed might not be the next 5 " [High,New] https://launchpad.net/bugs/726603 [15:45] klattimer, i marked it as High because people might trust that upcoming appointments are actually being displayed :) [15:45] klattimer, i missed a call myself... love brand new code :) [15:45] :P [15:48] rodrigo_, same on the other branch... just add the sanity-check against NULL... see comment -> https://code.launchpad.net/~rodrigo-moya/unity/ref-state-set-for-panel-service/+merge/50955 [15:48] rodrigo_, but approved [15:48] MacSlow, ok thanks! [15:53] apinheiro_afk: thing to ping us please when a branch is merged in trunk in case the milestone isn't set: see https://bugs.launchpad.net/unity/+bug/723806 [15:53] that helps tracking a lot :) === apinheiro_afk is now known as apinheiro [15:55] didrocks, checking ... [15:56] didrocks, ups, yes, you already told that the last week [15:56] sorry [15:57] no worry, just a kind reminder :) [15:57] didrocks, so this also applies to bugs 723804 and 722660 [15:57] Launchpad bug 723804 in unity (Ubuntu) "Require to implement AtkSelection on the Launcher" [Undecided,In progress] https://launchpad.net/bugs/723804 [15:57] right? [15:57] Launchpad bug 722660 in unity (Ubuntu) "Keyboard navigation: no public API to know the current Laucher Icon selected when key nav is activated" [Undecided,In progress] https://launchpad.net/bugs/722660 [15:57] apinheiro: right, targeting then :) [15:58] didrocks, can you set the milestone for today's release to https://bugs.launchpad.net/unity/+bug/715299 please? [15:58] sure :) [15:58] merci beaucoup [15:58] rodrigo_: it was already done [15:58] mais de rien :) [15:58] ah, cool [15:58] thanks apinheiro and rodrigo_! [15:59] didrocks, to you, that made the work [15:59] rodrigo_: basically, if it was already milestoned and we missed the target, we report it for next release automatically [16:00] ok [16:00] when is the release going to be done? [16:00] I have a branch almost done for a couple more bugs, although I need some time to test [16:05] rodrigo_: in 30 minutes approx. [16:06] ok, I'll leave it then for next release then [16:07] didrocks, what would be the version? [16:07] seb128: not an indicator datetime bug - https://bugs.launchpad.net/ubuntu/+source/indicator-datetime/+bug/699655 [16:07] klattimer, reassing it then? [16:08] seb128: wasn't assigned to me [16:08] just noticed your comment and added my own to clarify [16:08] move it to "gnome-panel" [16:08] k [16:08] what the hell, scrolling in qt creator isn't working on 1 file [16:08] right, that's what I meant by reassign, move it to the right source [16:11] klattimer, ok, i think it isn't necessarily startdate in the recurring ones [16:11] i think the sort is just broken [16:11] njpatel: is today dash day? [16:11] klattimer, i bumped it up to display 20 instead of 5 [16:11] jcastro, today dash day [16:12] kenvandine: well, that's a strange bug then :/ [16:12] and the order is all over the place, fri, tues, fri, mon, wed, fri, mon [16:12] hmm, that _IS_ weird [16:12] maybe something went wrong with my sort :/ [16:12] that is what i am thinking [16:12] a3Dman: look at http://launchpad.net/unity, the milestone says 3.6 :) [16:12] I can't see how [16:12] unless something weird happened with the lists or something :/ [16:12] I'll try and fix it [16:12] thx [16:12] i think it has to be that [16:13] clearly e_cal_component_get_dtstart [16:13] clearly e_cal_component_get_dtstart is returning the correct time, it is displayed right [16:13] jcastro, http://people.canonical.com/~njpatel/Screenshot-2.png [16:13] just broken sort [16:14] nice [16:14] I see multi monitor got sorted too? [16:14] somewhat [16:14] I'm going to try and fix it up better now [16:15] klattimer, i haven't completely ruled out recurring appointments being a factor... since most of mine are [16:15] kenvandine: ok [16:15] klattimer, perhaps you can test that if you don't have many recurring ones now [16:15] add a bunch or something [16:15] I'll see what I can work out [16:15] thx! [16:25] hmm, anyone has an example of running glib-genmarshal (or any other command that generates files) in cmake? I've tried a couple of things with no success [16:28] didrocks, there seems to be a regression in the auto-hide behavior of the launcher. a few weeks ago you fixed it so that when you press and hold the super key the launcher shows and as soon as the hand is lifted from the key launcher hides, its not the case anymore. now the launcher stays for a second and then hides [16:30] om26er: right, I think I commented that somewhere. As now there is the super + click to hide the launcher, we rewrapped this decision [16:30] it's on a bug, but part of deep archeology seeking :) [16:30] kenvandine: did you ever get the python stuff working with quicklists and libunity? [16:31] not quicklists [16:31] jcastro: it doesn't, there is still some work to do on an override [16:31] but it wouldn't work in vala either [16:31] ok [16:31] didrocks, ^^ [16:31] but mikkel knows right? [16:31] I packaged the first part, pitti will finish some fixes :) [16:31] anyone has a link for 64bit Natty devel snapshot? [16:31] jcastro: right, I worked on that with him last week [16:31] nmarques: http://cdimage.ubuntu.com/daily-live/current/ [16:31] kenvandine: the quicklist part is still not plugged btw [16:31] kenvandine: I will work on that after alt + f2 [16:31] * nmarques bows before jcastro [16:32] ok :) === om26er__ is now known as om26er [16:38] kenvandine: I think I might have found a slight hole which could be causing it [16:38] if I push a branch will you test and grab the debug output for me and attach it to the bug you filed? [16:38] sure [16:40] apinheiro, rodrigo_: well done for the upload, i'll let you guys mark the work items done now in https://launchpad.net/ubuntu/+spec/packageselection-dx-n-unity-a11y [16:40] dbarth, there are still time for a little addition in this upload? [16:41] mpt: hey [16:42] mpt: with your designer and HIG hat, can you have a look at https://bugs.launchpad.net/ubuntu/+source/compiz/+bug/507964 please? [16:42] didrocks, sure [16:43] (also, that can be linked with the https://bugs.launchpad.net/unity/+bug/723273 discussion) [16:44] didrocks, that's more of a JohnLea question [16:44] mpt: for the second one or the first one? [16:44] I don't know whether Unity even has a "desktop" to switch between [16:44] It doesn't have multiple panels to switch between. [16:45] first one [16:45] mpt: right, but the compiz default affects both sessions [16:45] (there is just one panel in unity, but we still have the two ones in the classic session) [16:49] DanRabbit, how are those icons coming along ? [16:49] might be disconnected in a minute [16:49] didrocks, heh [16:51] didrocks, I have some problems after your last commits [16:51] using Alt+f1 Im not able to give the focus to the launcher [16:51] kenvandine: lp:~karl-qdh/indicator-datetime/sort-debugging [16:51] thx [16:51] kenvandine: what we're looking for is "E-Cal Component is neither an event or a todo" [16:51] it is set on nautilus (in my case) [16:52] because if that's the case it'll tend to sort them to the end [16:53] apinheiro: can you get a test case? I'm using it right now and alt + F1 set it to the launcher here [16:53] didrocks, hmm, I need a nux update? [16:54] I did that on the morning [16:54] 11:00 am more or less [16:54] apinheiro: yeah, latest nux, but unity should even not built without it [16:54] ok, no, should be ok [16:54] didrocks, I will try it again [16:54] apinheiro: what do you have exactly, is the launcher hidden? [16:54] no, it is visible [16:54] apinheiro: so you are with a aways visible launcher? [16:55] or dodge window? [16:55] (default) [16:55] with the default I guess [16:55] I mean that if you are on a desktop screen [16:55] without any app running the launcher is being shown [16:55] right? [16:55] so, right [16:55] in this case [16:55] then, I press alt + F1 [16:56] but I will test it again [16:56] I see a launcher item (latest one getting the focus) having it [17:00] klattimer, no instances of "E-Cal Component is neither an event or a todo" printed from the service [17:00] kenvandine: well that's both interesting and worrying [17:00] hehe [17:00] interesting because I'd have assumed that's what caused it [17:01] let me add a else there to make sure things are cool [17:01] worrying because it means that there's something wrong with the sorting which doesn't make any particular sense :/ [17:02] apinheiro: you're right, something changed in trunk :/ [17:03] didrocks, Im right about what? [17:03] I require a nux update? [17:03] apinheiro: the focus doesn't work anymore, not sure what changed in trunk :/ [17:03] klattimer, added an else there and got lots of these [17:03] (process:7239): Indicator-Datetime-DEBUG: E-Cal Component is an event or a todo [17:04] so it is an event or todo :/ [17:04] apinheiro: additions? don't think so, the rest will be tracked with more bugs [17:04] didrocks: oh crap, it's breaking the uplaod? [17:04] dbarth: I'm looking at it [17:05] didrocks, take into account that commit 895 also modified some things related with [17:05] the laucncher [17:05] and the keynavmode [17:05] apinheiro: thanks for the pointer [17:05] in fact, I'm not sure why your changes are required [17:05]