=== doctormo__ is now known as doctormo === chaotic_ is now known as chaotic [09:54] * thorwil hands mpt a bonus point for using car analogy [10:04] mpt: regarding the menu problem being solved for submenus in gtk: afaics gtk does it time-based, wich means there's an annoying delay when browsing through menus. the proper way to do it would be direction based, instead [10:05] hrmpf, design.canonical.com seems to lack links to specific comments [10:07] thorwil, Windows is time-based, GTK is direction-based. [10:08] In Windows if you move diagonally towards the bottom of a submenu you'll get flash-flash-flash of items that aren't the submenu title. In GTK you won't. [10:08] mpt: hmm, then why is there such an annyoing delay before submenus show/hide? [10:08] thorwil, I reported a bug yesterday about the lack of comment permalinks [10:09] thorwil, that's an unrelated delay, to avoid submenus flashing open when you're moving past them on the way to lower items. [10:14] i really wish i could turn that off [10:20] ah, compiz was to blame, in part. the remaining delay is irregular and thus looks like a performance issue :/ [10:20] bbl === MacSlow is now known as MacSlow|lunch === vish is now known as mac_v === mac_v is now known as vish === MacSlow is now known as MacSlow|capoeira [17:45] A user reports in bug 568262 that Transmission and Rhythmbox disappear from Indicator Application after a while. How are we ever going to find out what caused this? [17:45] Launchpad bug 568262 in indicator-application (Ubuntu) "programs disappear from the applet (affects: 1) (heat: 6)" [Undecided,New] https://launchpad.net/bugs/568262 [18:02] qense, ask for .xsession-errors and .cache logs? [18:06] seb128: ah! Log files. Ingenious solution! [18:10] Sometimes I'm just not that sharp. :) === MacSlow|capoeira is now known as MacSlow [21:59] tedg, bratsche: So I think I finally narrowed down a possibility of what is happening using my small example. In libappindicator, you connect to the add signal for menus to detect when items are added to a menu. But, for some reason, that signal is (or seems to be) ever emitted when you add those submenus via gtk_ui_manager_add [22:02] Nafai: Hmm, I wonder if it's similar to the GtkContainer vs. MenuShell bug. [22:02] what's that one? [22:02] I think that it depended on how you added the entires they'd get picked up. Using one interface vs. the other. [22:03] One didn't emit any signals at all. [22:03] I have a minimal example that reproduces it [22:03] Plus added a few debug prints to lib app indicator [22:03] Nafai: Ah, cool. Sounds like something that should get assigned to bratsche though -- that's his code :) [22:04] ok, I'll assign it to him, with the details I just found out [22:04] bbiab, short errand [22:18] Nafai: Send me the link [22:37] https://bugs.launchpad.net/ubuntu/+source/gnome-bluetooth/+bug/558841 [22:37] Launchpad bug 558841 in gnome-bluetooth (Ubuntu) "bluetooth "devices" menu item not working in bluetooth indicator (affects: 3) (heat: 18)" [Medium,In progress] [22:38] The branch with my minimal example is linked there in the comments, I'm about to finish up a comment explaining what I believe is going on [22:40] Okay cool, thanks. [22:48] Nafai: Is minimal_menu_example supposed to do something other than crash? [22:49] uh, yeah :) [22:49] * bratsche installs, maybe that's the issue [22:49] actually [22:49] run it from the main directory [22:49] src/minimal_menu_example [22:49] Ah there we go [22:49] thanks. [22:49] I think the path to the .ui file was hardcoded :) [22:50] Cool it's running now. [22:50] I see: [22:50] Minimal [22:50] ------- [22:50] Quit [22:51] Right, if you look at the terminal output, I display the menu structure and the ui xml [22:51] * bratsche reads the code to see what it should be doing [22:51] Yeah [22:52] There should be a submenu on it right? [22:53] yes [22:53] and the add signal never seems to be signaled from the top-level menu [22:53] er, emitted [22:53] Nafai: Nice test program. This is great. :) [22:53] thanks [22:53] I tried to distill down what gnome-bluetooth was doing [22:56] Ah, I think I know exactly what's going on. [22:57] app-indicator.c:1277 [22:57] if (GTK_IS_TEAROFF_MENU_ITEM(widget)) { [22:57] return; [22:57] } [22:59] Eh, maybe that's not it. [23:00] Yeah, I had a debug print before that and I still didn't get the container_iterate called when it was added