[01:01] eeejay: see my merge request? [02:14] oh brother [02:15] hm? [02:15] this channel is not registered [02:17] vorian: How do we register it? [02:17] that's gonna be tricky :) [02:17] * tedg is relatively new to this stuff. [02:18] tedg: first things is this, a group contact needs to be established [02:18] * vorian gets another linky [02:18] http://freenode.net/group_registration.shtml [02:20] vorian: Okay, that makes sense. We're closely related to the Ubuntu folks. Does it make sense to put us under their contact, or to create another? [02:20] vorian: Or does that imply a namespace? [02:20] i would create another, but it is up to you all, and the #ubuntu IRC council [02:21] yeah, that would imply a namespace [02:21] * vorian would stick with this one [02:23] vorian: Thanks for the information, I think that davidbarth would probably be a good contact, but he's asleep right now. [02:23] no problem [02:23] vorian: I'll send him a mail, and hopefully get this started tomorrow. [02:23] excellent [02:24] vorian: What should we do about #dx? Wait until this is registered? [02:24] there's not much that can be done about it [02:24] Can you set the topic to "everyone's moved to #ayatana" ? [02:24] So atleast people know we're not leaving them behind :) [02:25] what if a nintendo fan comes in, and thinks "cool, we moved!" [02:25] Then we'll work on making them a Ubuntu fan ;) [02:25] :) [02:26] Perhaps text like "If you're looking for the Ubuntu DX channel it is now #ayatana" === glitch is now known as glitch- === glitch- is now known as glitch [09:19] bratsche, notification-daemon treats title text verbatim too [09:20] Try it: notify-send "Joe's phone is online" "fish & chips" [09:21] Compare with notify-send "Joe's phone is online" "fish & chips" [09:21] notification-daemon renders the "'" as "'" [09:21] so, so should notify-osd. [11:59] eeejay: hi eitan [12:01] do you have the latest pidgin package (pidgin + indicator stuff)? [12:24] * MacSlow -> lunch [14:20] hey tedg! [14:20] Good morning eeejay [14:21] tedg: kenvandine_wk: we need some quick decisions about the python bindings, eh? [14:21] eeejay: Yes. [14:22] kenvandine_wk: Did you get an initial upload sponsored? [14:22] yes [14:22] with the patch [14:22] kenvandine_wk: patch that lets it build against 0.1.5? [14:23] yes [14:23] ah, cool [14:23] eeejay: tedg has a branch that fixes some stuff that broke with 0.1.5 [14:23] eeejay: and i have a merge request too [14:23] think you can bring all that together and make a 0.0.3? [14:23] kenvandine_wk: you bet.. [14:24] eeejay: rock! [14:24] thx [14:25] kenvandine_wk: so this branch add some important dist files, bumps the version.. [14:25] * eeejay looks [14:25] yeah [14:25] the last tarball had 0.0.1 in configure.ac [14:25] but was a 0.0.2 tarball [14:25] hehe [14:26] also the COPYING files... we had some complaints in getting it reviewed about that [14:26] since it is dual licensed [14:26] so i did that [14:26] as well as the files in tests [14:26] kenvandine_wk: where does the build patch you made live? [14:26] tedg's branch [14:26] one sec [14:26] * eeejay didn't know it is dual licensed [14:27] it is in the source files :) [14:27] i assumed you did that [14:27] lp:~ted-gould/indicator-applet/python-bindings-dont-know-what-im-doing [14:27] kenvandine_wk: doh! [14:27] eeejay: We had to do LGPL v2/3 because some apps can't link against one or the other. [14:28] kenvandine_wk: tedg: ah.. phew, i thought i was just being sloppy [14:28] hehe [14:28] eeejay: i think merging my branch and tedgs into your's... then create a tarball [14:28] tedg: i couldn't get your branch to work with listen-and-print.py, is it supposed to? [14:28] eeejay: it would also be great if the tests files got installed as docs [14:29] eeejay: I didn't try that. I asked kenvandine_wk to check out Gwibber and Gajim. [14:29] eeejay: Probably I broke the listener interface. [14:29] tedg: but they don't use the listener iface [14:29] I said in the branch name that I didn't know what I was doing ;) [14:30] tedg: true, i guess you aren't liable then [14:30] i tested the im example, gwibber and gajim [14:30] kenvandine_wk: i updated trunk to work with 0.1.5 [14:30] kenvandine_wk: without tedg's branch [14:30] gajim actually only half works... but i think that is because the gajim guys changed my patch :) [14:30] eeejay: Do you see an easy thing that I broke? [14:31] eeejay: ok... i tested it with tedg's changes [14:31] tedg: you didn't break anything that was not broken already [14:31] eeejay: I was just trying to change to O instead of s and i for the various indicator struct pointers. [14:31] tedg: but you failed to fix the listener iface :) [14:32] Does it make sense to mark it as broken for 0.0.3 then? [14:32] tedg: yeah, that won't work. you obviously didn't know what you were doing ;) [14:32] tedg: yes, it does [14:32] * eeejay wrote about it extensively in the merge request [14:33] https://code.launchpad.net/~ted-gould/indicator-applet/python-bindings-dont-know-what-im-doing/+merge/5366 [14:34] eeejay: Heh, hadn't gotten to that mail yet. [14:34] eeejay: Yes, everything you say there :) [14:36] eeejay: hummm... it did work though :) [14:36] kenvandine_wk: what worked? [14:37] eeejay: So without the boxing, could we have something reasonable that would link against 0.1.5? It seems so if we drop listener, right? [14:37] tedg's code [14:37] tedg: exactly [14:37] And the fact that you manually undid the enum, that would be "unpythonic" but work. [14:37] kenvandine_wk: with gwibber and gajim? yeah, it should work with those [14:37] kenvandine_wk: the listener interface is broken [14:38] eeejay: ok... so it works fine for everything we know about now [14:38] kenvandine_wk: which only matters if you are implementing a new indicator-applet-like app [14:38] so not "awefule" [14:38] aweful even [14:38] but we should fix it [14:38] does anybody know if Bug #345837 is reallly a duplicate of Bug #345296? [14:38] tedg: the enum actually is not relevant, since the listener iface is broken [14:38] their descriptions seem to show that the bugs really are different. [14:39] eeejay: ok... so you already have these issues fixed? [14:39] kenvandine_wk: yeah, we should keep the gwibber and gajim folks happy [14:39] eeejay: Except for the signal coming from the server to say that someone is interested, right? [14:39] eeejay: I think, interested_added and interest_removed both use the enum. [14:40] tedg: which symbol does that? [14:40] I doubt that Gwibber or Gajim are using that though. [14:40] * eeejay looks at server.h [14:40] eeejay: today is freeze... so if we are getting a fix in, lets do it real soon :) [14:40] kenvandine_wk: okie dokes [14:41] hyperair: Seems odd to me too. MacSlow do you know if those bugs are dups ^^ [14:41] tedg: it would be nice if that oen could get fixed before jaunty, because it's really annoying if you use media keys with banshee. [14:42] So, to summarize. If we go with my "don't know what I'm doing branch" and make a 0.0.3 we'll link against 0.1.5 but the listener interface will be broken. And we've got a plan to make that better for 0.0.4. Is that a good summary eeejay kenvandine_wk? [14:42] * eeejay moves swiftly [14:43] tedg: the relevant changes from your branch have been in trunk for a while [14:43] * eeejay acts swiftly [14:43] eeejay: well how hard is fixing the listener interface? [14:43] eeejay: Heh, okay. Are you going to spin a 0.0.3 then or should I? [14:43] it would be nice to make a 0.0.3 with that [14:44] tedg: cause what we have now is as good as that... no need to spin a release :) [14:44] tedg: let me get the branch in shape, we could all agree, and i will cut a release [14:44] tedg: 0.0.2 was uploaded with the patch from your branch [14:45] kenvandine_wk: Yeah, I'd like to get something cleaner than what I did if possible. [14:45] tedg: yeah... but i am hoping we can fix the other issue at the same time :) [14:45] kenvandine_wk: I'm not a python fan, and I think it shows :) [14:45] hehe [14:45] * kenvandine_wk would chose python over C any day... but lets not have that debate [14:49] * tedg is going to make a T-shirt: "You can take my semicolon from my cold dead hands." :) [14:52] * eeejay rejects tedg's branch because of his hateful comments [14:53] * hyperair kills tedg and takes his semicolon [14:53] kenvandine_wk: would you mind testing trunk one last time before i upload the tarball? [14:53] tedg: i believe it's s/take/pry/ [14:54] eeejay: certainly [14:54] hyperair: Heh, need to spend more time at NRA rallies... research... [14:54] eeejay: is it ready now? [14:54] kenvandine_wk: yup! [14:55] kenvandine_wk: lp:~indicator-applet-developers/indicator-applet/indicate-python <- to avoid confusion [14:56] tedg: ? [14:57] hyperair: I believe that the guns phrase that is taken from is an NRA one. [14:57] testing nwo [14:58] i think charlton heston would say that at NRA rallies [14:58] tedg: no idea. i see it everywhere [14:58] or michael moore lead me to believe that [15:04] kenvandine_wk: not so smooth? [15:06] gah. Did this latest version of pidgin stuff get tested? [15:06] i'm now getting pidgin segfaulting. [15:06] eeejay: sorry... phone rang [15:06] testing now [15:06] kenvandine_wk: phew, thought you were in debugging craze [15:10] Hobbsee: pidgin-libnotify? There is a big pidgin bug with Jabber right now. [15:10] eeejay: not good [15:10] kenvandine_wk: crasher? [15:10] tedg: oh, there was another change there. I assumed it was notify killing it. [15:11] File "/usr/lib/python2.6/dist-packages/indicate/__init__.py", line 26, in [15:11] from _indicate import * [15:11] ImportError: /usr/lib/python2.6/dist-packages/indicate/_indicate.so: undefined symbol: indicate_server_show_indicator_to_user [15:11] Hobbsee: It seems to be related to gstreamer -- I don't really get how though. [15:11] eeejay: ^^ [15:12] tedg: oh, fun. [15:12] kenvandine_wk: checking now.. [15:13] tedg: thanks. Wish I knew how to disable the jabber bit! [15:13] Hobbsee: I don't think there's a way, but if you uninstall one of the gstreamer plugin packs, I think bad, it gets better. [15:14] tedg: thanks [15:16] kenvandine_wk: i don't have that symbol in my version, are you certain you installed it in /usr/lib/python2.6/dist-packages/indicate? [15:16] yes [15:16] i did a make dist [15:16] then build the package from that tarball [15:16] let me compare files [15:17] kenvandine_wk: if you didn't give it /usr as the prefix, it should be in /usr/local [15:17] oh [15:17] humm [15:17] eeejay: i installed it from a package [15:17] it didn't use my tarball [15:17] hang on [15:18] you scared me for a moment [15:18] i usually give it some arbitrary prefix, and set PYTHONPATH to that [15:18] i like testing with packages [15:18] i am a nut that way [15:19] heh [15:20] davidbarth: you marked bug #345837 as a dupicate of #345296 right? are you sure it's really a duplicate? [15:20] davidbarth: the descriptions seem rather different. [15:22] hyperair: hi, let me check again [15:23] davidbarth: alright [15:24] hyperair: you're right this one is not a duplicate [15:25] it is the issue with the close_notification_handler [15:25] hmm [15:25] haven't looked at notify-osd's code yet [15:25] tedg: speaking of pidgin, has anyone tested the notifications on dist-upgraded machines? I found some strangeness, before i wiped my ~/.purple directory, and started again [15:25] but it would be good if the priority for this bug could be raised a little [15:25] but there was anohter bug I thought with banshee/rhythmbox bubbles crashing into synchronous ones [15:25] it's very annoying, and very visible when you use banshee with media keys [15:26] hyperair: the problem has been fixed (sort of) by not sending confirmation bubbles from gnome-settings-daemon [15:26] davidbarth: rhythmbox doesnt have this issue, because it updates the existing bubble rather than deleting the old and spawning a new [15:26] hyperair: from an architecture point of view, that was not great [15:26] hmm. why not? [15:26] so the bug should not occur anymore [15:26] even if the corner-case still exists in the layout code of n-osd [15:26] davidbarth: it won't occur as often, but it'll still occur. [15:26] Hobbsee: I haven't tried that specifically. Though I haven't noticed any bugs related to something like that. What was the weirdness? [15:27] davidbarth: e.g. when i adjust my volume when pidgin fires up a notification [15:27] davidbarth: and there are multiple [15:27] hyperair: ah, and pidgin /force/ closes the bubbles? [15:27] eeejay: ok... it works [15:27] the listener still shouldn't work right? [15:27] tedg: new messages in pidgin not showing up in the envelope at all (although it would show up most of the time with the little black notification) [15:27] kenvandine_wk: sweet! [15:28] davidbarth: you don't need to force close the bubbles for it to happen i think [15:28] kenvandine_wk: right [15:28] kenvandine_wk: pushing a tarball now [15:28] thx [15:28] davidbarth: but okay, supposing you change your volume while banshee's switching songs, you'd get that. [15:28] tedg: ie, if you missed the black notification, there was absolutely no indication of a new message [15:28] hyperair: the bug with update was fixed, as well as other potential visual collisions between sync and async bubbles [15:29] Hobbsee: That might have been related to a couple of bugs that we just fixed in indicator-messages, they had the possibility of getting lost. [15:29] it's in the lates tarball of today (not yesterday) [15:29] brb [15:29] Hobbsee: Hopefully those are fixed, I'm curious why removing .purple would solve that. [15:29] Hobbsee: Honestly, I'm hoping they were unrelated :) [15:30] tedg: I'm not sure. I tested it a few days ago, removed ~/.purple, set up the user accounts again, and it worked fine. [15:30] davidbarth: i see, then it should be considered fixed then =\ [15:30] tedg: might be worth trying to find some people to test that [15:33] tedg: i don't think i have perms to register a new release in the 0.1 indicator-applet series [15:34] Oh nice, I think this new text filter works well. [15:35] eeejay: Okay, made it: https://edge.launchpad.net/indicator-applet/0.1/python0.0.3 [15:35] eeejay: Can you add files? [15:35] tedg: yep, thanks [15:38] tedg: kenvandine_wk: released [15:38] eeejay: Yeah! Great! [15:39] great [15:39] i already subscribed motu-release :) [15:39] bug 358483 [15:40] sweet [15:41] asac and james_w are looking at it [15:43] eeejay: mind filing a bug about the listener so we can track it? [15:43] kenvandine_wk: yup, will do [15:43] thx [15:43] please subscribe me too :) [19:16] away [19:17] whoops === rickspencer3 is now known as rickspencer3-afk === rickspencer3-afk is now known as rickspencer3